Full text of "9118"
Energy in Developing Countries
January 1991
OTA-E-486
NTIS order #PB9 1-133694
ENERGY IN DEVELOPING COUNTRIES
CONGRESS OF THE UNITED STATES
OFFICE OF TECHNOLOGY ASSESSMENT
Recommended Citation:
U.S. Congress, Office of Technology Assessment, Energy in Developing Countries,
OTA-E-486 (Washington, DC: U.S. Government Printing Office, J anuary 1991).
For sale by the Superintendent of Documents
U.S. Government Printing Office, Washington, DC 20402-9325
(order form can be found in the back of this report).
Foreword
This report was prepared in the course of the ongoing OTA Assessment, Fueling
Development: Energy and Technology in Developing Countries, which is being carried out in
response to requests from the Senate Committee on Governmental Affairs; the House
Committee on Energy and Commerce; the Subcommittee on Energy and Power of the House
Committee on Energy and Commerce; the Subcommittee on Human Rights and International
Organizations and the Subcommittee on Africa of the House Committee on Foreign Affairs;
the Subcommittee on International Development, Finance, Trade and Monetary Policy of the
House Banking Committee; and individual members of the Senate Environment and Public
Works Committee; the House Select Committee on Hunger; and the Congressional
Competitiveness Caucus.
These committees and subcommittees asked OTA to examine the role of technology in
providing cost effective energy services that poor countries need for their economic and social
development, while minimizing adverse environmental impacts. The committees were
particularly interested in an analysis of the U.S. role in facilitating the adoption of such
technologies. The overall assessment is scheduled for completion in 1991. This report, the first
of two, was prepared in response to the requesting committees' interest in receiving an interim
product. It examines how energy is supplied and used in developing countries, and how energy
use is linked with economic and social development and environmental quality.
OTA received substantial help from many organizations and individuals in the course of
preparing this report. We are very grateful for the efforts of the project's contractors, who
prepared parts of the background analysis; members of the advisory panel; and workshop
participants, who provided guidance and extensive critical reviews; and the many additional
reviewers who gave their time to ensure the accuracy of this analysis.
U JOHN H. GIBBONS
Director
Advisory Panel — Energy and Technology in Developing Countries
Harry G. Barnes, J r., Chairman
Critical Languages and Area Studies Consortium
Irma Adelman
Department of Economics
University of California, Berkeley
J effrey Anderson
Institute of International Finance
Edward S. Ayensu
Pan-African Union for Science and Technology
Gerald Barnes
General Motors Corp.
Mohamed T. El-Ashry
World Resources Institute
Eric Helland-Hansen
United Nations Development
Programme
Carl N. Hodges
Environmental Research Laboratory
University of Arizona
Donald J est
Sun Refining and Marketing Co.
Amory Lovins
Rocky Mountain Institute
Mohan Munasinghe
World Bank
Henry Norman
Volunteers In Technical Assistance
Waafas Ofosu-Amaah
WorldWIDE
R.K. Pachauri
Tata Energy Research Institute
D. Lou Peoples
Bechtel Power Corp.
Gabriel Sanchez-Sierra
Organisacion Latino-Americana de Energia
Kirk R. Smith
Environment and Policy Institute
East-West Center
Irving Snyder
Dow Chemical U.S.A.
Thomas B. Steel, J r.
Natural Resources Defense Council
Robert H. Williams
Center for Energy and Environmental Studies
Princeton University
Lu Yingzhong
Professional Analysis Inc. (PAI)
Montague Yudelman
World Wildlife Fund and The Conservation
Foundation
Eugene W. Zeltmann
General Electric Co.
Observer
David J hirad
U.S. Agency for International Development
NOTE: OTA appreciates and is grateful for the valuable assistance and thoughtful critiques provided by the advisory panel members.
The panel does not, however, necessarily approve, disapprove, or endorse this report. OTA assumes full responsibility for the
report and the accuracy of its contents.
OTA Project Staff-Energy in Developing Countries
Lionel S. J ohns, Assistant Director, OTA
Energy, Materials, and International Security Division
Peter D. Blair, Energy and Materials Program Manager
Project Staff
Joy Dunkerley, Project Director
Samuel F. Baldwin, Senior Analyst
Karen Larsen, Senior Analyst
Robin Roy, Senior Analyst Paul S. Komor, Analyst
Nina Goldman, Research Assistant Sharon Burke, Research Assistant
Roger Chin, Summer Intern
Administrative Staff
Tina Brumfi eld Lillian Chapman Linda Long
Contributors
Rosina Bierbaum, OTA Oceans and Environment Program
Gretchen Kolsrud, OTA Industry, Technology, and Employment Program
Howard Levenson, OTA Oceans and Environment Program
Edward MacDonald, OTA Food and Renewable Resources Program
Contractors
Russell deLucia
deLucia and Associates, Inc.
Cambridge, MA
Ashok Desai
National Council of Applied Economic Research
New Delhi, India
K.G. Duleep and Sergio Ostria
Energy and Environmental Analysis, Inc.
Arlington, VA
Ahmad Faruqui and Greg Wikler
Barakat and Chamberlain, Inc.
Oakland, CA
Howard Geller
American Council for an Energy Efficient
Economy
Washington, DC
Theodore J . Gorton
Petroleum Consultant
Bethesda, MD
Donald Hertzmark
Independent Consultant
Washington, DC
Virendra Kothari
Energy and Environmental Analysis, Inc.
Arlington, VA
Karin Lissakers
Columbia University
New York, NY
Arjun Makhijani
Institute for Energy and Environmental Research
Takoma Park, MD
Peter Meier and J ohn Lee
International Development and Energy
Associates, Inc.
Washington, DC
Henry Peskin
Edgevale Associates, Inc.
Silver Spring, MD
Vaclav Smil
University of Manitoba
Winnipeg, Canada
Environmental Problems and Priorities in Developing Countries
Workshop Participants, April 19, 1990
Michael Adler
U.S. Environmental Protection Agency
Lutz Baehr
United Nations
Center for Science and Technology in Development
Leonard Berry
Florida Atlantic University
Al Binger
Conservation Foundation/Biomass Users Network
J a net Welsh Brown
World Resources Institute
Lalanath de Silva
Environmental Foundation, Ltd.
Sri Lanka
Clarence Dias
International Center for Law in Development
Paul Dulin
Associates in Rural Development
J ohn J . Gaudet
U.S. Agency for International Development
Robert Goodland
World Bank
Lupe Guinand
BIOMA
Venezuela
Robert I chord
U.S. Agency for International Development
Kari Keipi
Inter-American Development Bank
Ananda Krishnan
United Nations
Center for Science and Technology in Development
Russell Mittermeier
Conservation International
Hind Sadek
WorldWIDE
Paul Schwengels
U.S. Environmental Protection Agency
Thomas B. Steel, J r.
Natural Resources Defense Council
NOTE: OTA is grateful for the valuable assistance and thoughtful critiques provided by the workshop participants. The views
expressed in this OTA report, however, are the sole responsibility of the Office of Technology Assessment.
Outside Reviewers
Irma Adelman
Department of Economics
University of California, Berkeley
Dilip Ahuja
Environmental Protection Agency
J effrey Anderson
Institute of international Finance
Edward S. Ayensu
Pan-African Union for Science and Technology
Harry G. Barnes, J r.
Critical Languages and Area Studies Consortium
Martin J . Bernard III
Argonne National Laboratory
Vic Fazio
Member of Congress
Robert Goodland
World Bank
Eric Helland-Hansen
United Nations Development Programme
Carl N. Hodges
Environmental Research Laboratory
University of Arizona
David J hirad
U.S. Agency for International Development
Donald J ones
Oak Ridge National Laboratory
Donald J est
Sun Refining and Marketing Co.
Dolores Kern
National Coal Association
Barbara Lausche
World Bank
Henry Norman
Volunteers In Technical Assistance
Waafas Ofosu-Amaah
WorldWIDE
Philip O'Keefe
ETC (U.K.)
R.K. Pachauri
Tata Energy Research Institute
Gabriel Sanchez-Sierra
Organisacion Latino-Americana de Energia
Anjali Sastry
Rocky Mountain Institute
J ayant Sathaye
Lawrence Berkeley Laboratory
Lee Schipper
Lawrence Berkeley Laboratory
Kirk R. Smith
Environment and Policy Institute
East-West Center
Irving Snyder
Dow Chemical U.S.A.
Thomas B. Steel, J r.
Natural Resources Defense Council
Carlos Suarez
Institute de Economia Energetica
Fundacion Bariloche
Pamela Wentworth
Bechtel Power Corp.
Thomas J . Wi I banks
Oak Ridge National Laboratory
Robert H. Williams
Center for Energy and Environmental Studies
Princeton University
Lu Yingzhong
Professional Analysis Inc. (PAI)
Montague Yudelman
World Wildlife Fund and The Conservation
Foundation
Eugene W. Zeltzmann
General Electric Co.
NOTE: OTA is grateful for the valuable assistance and thoughtful critiques provided by the reviewers. The VI eWS eXpreSSQ 1 1") this
OTA report, however, are the sole responsibility of the Office of Technology Assessment.
Contents
Chapter Page
Chapter 1: I introduction and Overview 3
Chapter 2: E nergy and Economic Development 25
Chapter 3: Energy Services in Developing Countries 47
Chapter 4: Energy Supplies in the Developing World 93
Chapter 5: Energy and the Environment in Developing Countries : Ill
Appendix A: Glossary of Energy Units 137
Wi
Chapter 1
Introduction and Overview
Contents
Page
I introduction 3
The Purpose of This Assessment 3
The Developing World and the Industrialized World , 4
Similarities and Differences Among Developing Countries 5
The Developing World in Global Energy 7
Analytic Focus 9
Overview of the Report 10
Population Growth 11
Economic Development . . * * 11
Energy Supply Constraints 12
Financial Constraints 12
Biomass Supply Constraints 13
Institutional Constraints on Rapid Expansion in Energy Supplies 14
Environmental Degradation in Developing Countries 15
Greenhouse Gases and Developing Countries 17
Prospects for Efficiency Improvements in Energy Production and Use 18
Appendix 1A: Economic, Social, and Energy Indicators for Developing Countries .... 20
Figures
Figure Page
l-i. Differences Between Developing and I ndustrial Nations 4
1-2.1985 Energy Consumption, Industrial and Developing Region Fuel Mix 8
1-3. Commercial Energy Consumption, 1973, 1985, and 2020 10
1-4. Per-Capita Commercial Energy Consumption, 1973, 1985, and 2020 10
1-5. Suspended Particulate Matter Levels in Selected Cities, 1980-84 15
1-6. Sulfur Dioxide Levels in Selected Cities, 1980-84 16
Tables
Table Page
1-1. Heterogeneity of the Developing World: Social, Economic,
and E nergy I ndicators 6
1-2. Commercial Energy I mport Dependence in Developing Countries 7
1-3.1985 Primary Energy Supplies 8
1-4. Largest E nergy Consumers, 1987 9
1-5. Passenger Fleet Annual Growth in Selected Countries 12
1-6. Electric Appliance Ownership in Urban Areas 12
1-7. Energy Imports, Debt Service, and Export Earnings for
Selected Developing Countries, 1987 13
1-8. Estimated Annual Energy Investment as a Percentage of Annual Total
Public Investment During the Early 1980s 14
Chapter 1
Introduction and Overview
Introduction
Energy use in developing countries has risen more
than fourfold over the past three decades and is
expected to continue increasing rapidly in the future.
The increase in the services that energy provides is
necessary and desirable, since energy services are
essential for economic growth, improved living
standards, and to provide for increased human
populations. But finding the energy supplies to
provide these services could cause major economic
and social problems. For many of the developing
countries, much of the additional energy needed will
be supplied by imported oil, and rising oil imports
will further burden those countries already saddled
with high oil import bills. Similarly, building dams
or powerplants to meet higher demands for electric-
ity could push these nations even deeper into debt.
Energy development and use also contribute to local
environmental damage in developing countries,
including record levels of air pollution in some
urban areas.
The rapid growth of energy use in developing
countries has wide impacts. The economic develop-
ment process has traditionally been accompanied by
rapid increases in oil demand, which, together with
rising demand in the industrial countries, contribute
to upward pressures on world oil prices. High levels
of indebtedness in the developing countries, partly
energy-related, have already contributed to instabil-
ity in the international money and banking system.
Rapid increases in fossil fuel use in developing
countries also represent a growing contribution to
the increase in local and regional air pollution as
well as atmospheric concentrations of greenhouse
gases such as carbon dioxide (C0 2 ). International
efforts to control greenhouse gas emissions require
active participation by developing countries. Many
developing countries could be adversely affected by
climate change, some much more than most indus-
trial nations.
An economically and environmentally sound
approach to energy development offers potentially
large benefits both for the developing countries and
for the rest of the world. It can contribute to
economic growth in the developing countries, lead-
ing to higher living standards, reduction of hunger
and poverty, and better environmental quality. This
strategy also holds benefits for the richer countries.
The developing countries are important trading
partners for the United States. More rapid economic
growth in these countries could stimulate U.S.
exports, including exports of energy technology
products, and, therefore, could benefit the U.S. trade
balance. Improved energy technologies can slow the
rate of increase in greenhouse gas emissions-a
global benefit.
The Purpose of This Assessment
This report is part of an assessment entitled
"Fueling Development: Energy and Technology in
the Developing Countries,' requested by the Senate
Committee on Governmental Affairs; the House
Committee on Energy and Commerce; the Subcom-
mittee on Energy and Power of the House Commit-
tee on Energy and Commerce; the Subcommittee on
Human Rights and International Organizations and
the Subcommittee on Africa of the House Commit-
tee on Foreign Affairs; the Subcommittee on Inter-
national Development, Finance, Trade and Mone-
tary Policy of the House Banking Committee; and
individual members of the Senate Environment and
Public Works Committee, the House Select Com-
mittee on Hunger, and the Congressional Competi-
tiveness Caucus.
The Office of Technology Assessment (OTA)
was asked to examine the extent to which technol-
ogy can provide the energy services that developing
countries need for economic and social development
in a cost-effective and socially viable manner, while
minimizing the adverse environmental impacts; and
to evaluate the role of the United States in accelerat-
ing the adoption of such technologies by developing
countries.
This report, the frost product of the assessment,
examines how energy is currently supplied and used
in the developing countries and how energy is linked
with economic and social development and the
quality of the environment. Our emphasis is primar-
ily on the present status of developing countries, and
concerns about current energy trends. This report is
intended to provide an introduction to the problems,
challenges, and opportunities associated with pro-
4 •Energy in Developing Countries
viding energy services for economic and social
development in the developing countries. These
issues are examined under four broad topic areas:
energy and economic development (ch. 2); energy
services (ch. 3); energy supplies (ch. 4); and energy
use and the environment (ch. 5). A subsequent report
will present the results of OTA's assessment of
technologies that can potentially improve the effi-
ciency of both energy production and use in develop-
ing countries; an examination of the technology
transfer process; and ways in which Congress can
help promote the rapid adoption of such policies.
The Developing World and the
Industrialized World
We largely follow the definition of "developing"
countries-low- and middle-income countries (fur-
ther divided into lower middle and upper middle
countries)— used by the World Bank 1 (see app. 1A
for a list of these countries), including all of the
countries of Africa, Latin America, and Asia,
excluding J apan. 2
There are wide differences in average indicators
of social and economic conditions between develop-
ing and industrial countries (figure l-l).
1. Social: The citizens of (OECD) countries have
a longer life expectancy (76 years, compared
with an average of 62 in developing countries),
largely due to lower infant mortality rates (9 per
1,000 live births compared with 71 in develop-
ing countries). A much larger share of the
Figure l-l — Differences Between Developing and
Industrial Nations (developing nation average
as a share of Industrial* nation average)
Real GDP
per capita
Commercial energy
consumption
per capita
Adult literacy
Life expectancy
0% 25% 50% 75% 100%.
How to Interpret this figure. Theaverage values for the industrial
countries are assigned 100 in all cases. The values for the
developing countries are expressed as a share of 100. For
example, average life expectancy in the industrial nations is 76
years and in the developing nations 62 years, or 82 percent of the
industrial country level.
"Industrial excludes the U. S. S. R.; based on weighted average of
high-income market economies.
SOURCE: United Nations, Human DevelopmenfReport 199a World Bank
WorldDevelopment Report.
population has access to secondary and higher
education and health care.
2. Economic: Average per-capita incomes 3 (ex-
pressed in purchasing power parities") are more
than eight times higher in OECD countries than
in the developing countries. This difference in
income levels reflects major differences in
economic structure, particularly the higher
share of agriculture in total production in the
developing countries. A much lower share of
lgee, for example, World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 1989), pp. 164-165.
2 The definition of developing countries is based primarily on per-capita income levels. This ranking system is rather arbitrary, however; if all the
countries of the world are ranked by ascending level of per-capita income, there is no obvious gap in the series to demarcate the two groups of countries.
The World Bank distinguishes six categories of counties: low-income, lower-middle-income, upper-middle-income, high-income oil-exporters,
industrial market economies, including OECD as a subset, and "non-reporting non-members" (the U.S.S.R., North Korea, East Germany, Angola,
Bulgaria, Albania, Mongolia, and Namibia). The group of developing countries (low- a nimiddle-income and upper-rniddle-income countries) does not
include Saudi Arabia, Kuwait, and the United Arab Emirates by virtue of their high per-capita income. The World Bank does, however, include as
developing countries some East and West European countries, such as Poland, Hungary, Yugoslavia, Greece, and Turkey, that qualify as developing
countries by virtue of their income levels, but, due to their integration with industrial economies of East and West Europe, do not share other
characteristics of underdevelopment, and are therefore not included in this report. Some other countries are excluded due to lack of reported data. Where
group averages of general economic and social indicators are reporteddirectly from theWorld Development Report, these countries are included in the
total. In more detailed analysis, they are excluded. While every effort is made to adhere to thesedefinitions, it is not always possible, especially when
other sources of data with slightly differentdefinitions are used.
'Income is usually measured by Gross Domestic or Gross National Product. The difference between the two — typically small for most countries — is
that GDP measures the total output of goods and services within the national border of a country, whereas GNPmeasures the output of goods andservices
attributable to the nationals of a country wherever that activity occurs.
4 If market exchange rates ate used to convert the GDP of different countries to dollars, averageOECD per-capita income appears to be over 20 times
higher than average developing country per-capita income. If, however, the comparison of income levels is adjusted to take in account differences in
purchasing power of currencies (i.e., what a unit of currency such as the dollar will buy in different countries) the gap between average per-capita income
levels in developing and OECD countries narrows, and OECD per-capita income levels are 8 times rather than 20 times higher tban the developing
country average. In either case, the gap in income levels is substantial.
Chapter 1 -Introduction and Overview «5
the population lives in urban areas-37 percent
in the developing countries compared with 77
percent in the OECD countries. Population
growth is more rapid in developing countries.
It is estimated to double by 2040, while the
population of the industrial world will increase
by only 15 percent over the same period.
3. Energy: The economic and social contrasts are
also reflected in energy consumption. Per-
capita consumption of commercial energy (coal,
oil, gas, and electricity) 5 in the OECD countries
is on average 10 times higher than in the
developing countries. On the other hand, com-
mercial energy consumption is increasing much
faster in the developing countries. Biomass
energy consumption in the developing coun-
tries is higher than in the OECD countries and
provides a much higher share of total energy
consumption.
Similarities and Differences Among
Developing Countries
The developing country averages shown in figure
1-1, though adequate to illustrate the broad contrasts
between developing and industrial countries, ob-
scure the wide economic and social differences
among developing countries. Indeed, the range of
differences between LDCs is greater than that
between many of them and the industrial countries.
A generation of exceptionally fast economic growth
in the Newly Industrialized Countries (the NICs),
combined with the slow growth, or in some cases,
economic stagnation and decline, in many African
countries, has widened the gap among developing
countries. Thus the problems, energy or otherwise,
faced by a relatively rich and developed country
such as Brazil are different from those faced by a
poor country like Ethiopia, as are the resources
available for their solution. An appreciation of these
differences is necessary for the realistic assessment
of energy technologies.
Per-capita incomes in the upper middle-income
developing countries (eg., Brazil, Argentina, Alge-
ria, Venezuela, and Korea) are almost seven times
higher than in the low-income countries (table 1-1). 6
The income differential reflects major differences in
economic structure. In the upper middle-income
countries, industry has a much larger share in total
output and agriculture a much lower share. I ndia and
China are exceptions, with atypically large shares of
industry, given their levels of income. The share of
the total population living in urban areas is much
lower in the low-income countries. For example, in
several African countries only about 10 percent of
the total population is urban'dwellers, in contrast to
countries like Brazil, Argentina, and Venezuela,
whose levels of urbanization (about 80 percent of the
population living in towns) are similar to those in the
industrial countries.
Developing countries also show wide variations
in social indicators. Life expectancy at birth rises
from an average of 54 years in the low-income
developing countries to an average of 67 in the upper
middle-income countries. Infant mortality is twice
as high in the low-income countries (over 100 per
1,000 births compared with an average of 50 in the
upper middle-income developing countries). India
and China are again exceptions: in both countries,
despite lower average income, indicators of social
development are similar to those found in countries
with much higher incomes. The experience of these
two countries testifies to the importance of social
policies in achieving relatively high levels of social
development despite low incomes. 8 Population
growth rates also differ widely among developing
countries. In recent years these have ranged from
about 1 percent annually in some countries (e.g.,
China, Uruguay, Korea) to over 3 percent in several
African countries.
The wide variations in social and economic
conditions in developing countries are also reflected
in their energy use. In the upper middle income de-
Theterm "commercial energy" conventionally applies to coal, oil, gas, and electricity on the basis that they are widely traded in organized markets.
These fuels are distinguished from other fuels such as firewood, charcoal, and animal and crop wastes, which are described as "biomass" or
"noncommercial fuels." The distinction between them can be misleading,particularly in the context of developing countries, as some of the so-called
"noncommercial" fuels, such as firewood and charcoal are also widely traded in highly organized markets. To minimize this ambiguity we use the term
"biomass energy" here.
*See also app. 1A.
Estimates of urban populations are based oncountry-specific criteria related to size of settlement and presence of urban characteristics.
8 This theme is developed further in a recent publication by the United NationsDevelopment Programme, Human Development Report 1990 (New
York, NY: Oxford University Press, 1990).
'See footnote 5.
On
£
3
3
Table 1-1 — Heterogeneity of the Developing World: Social, Economic, and Energy Indicators
Low-income India and Lower middle- Upper middle-
Indicators Year countries China income countries income countries
Economic:
GNP per capita ($1987) (ppp) 1 1987 840.0 900.0 3,000.0 5,420.0
Share agriculture in GDP (%)' 1987* 33.0 30.0 21.0 10.O*
Urban population as share of total (%) c 1987 24.0 33.0 51.0 66.0
Cars and trucks per 1,000 members of Population" 1980* 3.2 3.2 19.3 93.3
Social:
Life expectancy at birth (years)' 1987 54.0 65.0 64.0 67.0
Infant mortality per 100 births' 1987 103.0 62.0 61.0 50.0
Share of age group with secondary education (%)9 1986 25.0 39.0 51.0 59.0
Energy:
Commercial energy consumption per capita
(gigajoules)" 1986 4.9 16.3 36.2 58.3
Total energy consumption per capita (gigajoules)' 1986 12.3 18.7 41.7 67.0**
Share of traditional energy in total (%} ' 1986 60.0 13.0 13.0 13.0**
* Estimated.
If Brazil is excluded, the total per capita energy consumption would be 62 gigajoules and the share of traditional energy in total 2 percent.
SOURCES: 'World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 1969). Purchasing power parity (ppp) estimate based on data on pp. 164 and 222.
"Ibid., based on data in table 3.
c lbid., table 31.
d JoY Dunkerley and Irving Hoch, Transport Energy: Determinants and Policy (Washington, DO: Resources for the Future, September 1985), table 5-1 and appendix table 13> Based on
estimates. Note that totals are unweighed averages and the countries included differ slightly from Work/ Development Report 1989.
e World Bank, op. cit, table 1.
Ibid., table 32.
g Ibid., table 29.
"Ibid, table 8.
'United Nations, Energy StatisticsYearbook 1986( New York, NY: United Nations, 1988), table 4.
These values for the share of traditional energy are muoh lower than those found in field surveys. These values are presented herein order to have a consistent data set. Estimates
based on field surveys suggest that biomass provides one-third of the energy used by developing countries overall (efts. 3 and 4).
to
I"
I'
s»
R
a
3
Chapter 1 --Introduction and Overview »7
Table 1-2-Commercial Energy Import Dependence in Developing Countries
Country Number of Number of Number of High Medium Low
income countries energy energy importers importers importers
group in group 1 exporters importers (70-100%)b (30-70%)" (0-30%)"
Low-income 38 4 34 29 3 2
China and India 2 1 1 1
Lower middle-income 30 10 20 15 3 2
Upper middle-income 10 6 4 2 1 1
Total 80 21 59 46 7 6
a lnd~es all countries for which import dependence data are available,
b shares of imports in total commercial energy Consumption.
SOURCE: Based on data in the United Nations, 1986 Energy Statistics ¥earbook(New York, NY: 1988).
veloping countries, per-capita annual commercial (cooking, subsistence agriculture). At the other end
energy consumption (at 60 gigajoules' ) is 12 times of the scale, energy use by the economically well off
higher than in the low-income countries (5 GJ )." is also reasonably similar between developing and
Again China and India differ from the other low- industrial countries, in terms of quantity used (to
income countries, with per-capita consumption of within a factor of 2 or 3), source (oil, gas, coal,
commercial energy more than 3 times higher than electricity), and services provided (electric lighting
other low-income countries. Per-capita consumption and appliances, industrial goods, private automo-
of traditional biomass fuels, on the other hand, is biles, etc.). The large differences between countries
generally higher in the poorest countries, depending are then in large part due to the relative share of the
on the biomass resources available 12 total traditional villagers and the economically well
T, ■■iii ■ ,■ off in the population, and in the forms and quantities
There are similarly large variations in energy rf J j^ ^^ ^ ^ ^h ^
resource endowment. While many countries have transition tetween these two QCtremeS- The broad
some energy resources three-quarters of the deveU s i m i| ar i tie5 within spet]f]c population sectors imply
oping countries depend on imports for part or all of that jt js jb|e to make generalizations about
their commercial energy supplies (table 1-2). Levels technology that are applicable to a wide range of
of import dependence vary, but in many countries dh ^ di te countries
imports (almost entirely oil) provide nearly all
commercial energy supplies. Oil imports can be a .
considerable strain on already tight foreign ex- Ahe Developing World in
change budgets. In several countries, particularly in Global Energy 13
Africa and Central America, oil imports represent , , . t .
over 30 percent of foreign exchange earnings from _ The developing countries now account for about
exports (see app 1A) 30 percent of global energy use, including both
commercial and traditional energy (see table 1-3),
Despite these differences in aggregate indicators, and their share is growing rapidly. Their use of the
there are strong similarities among developing different fuels vary widely: they account for 85
countries within specific sectors. Energy use in percent of biomass fuel consumption but only 23
traditional villages throughout the developing world percent of commercial fuels (oil, gas, coal, and
is fairly similar in terms of quantity used, source electricity). The main sources of energy for the
(biomass, muscle power), and services provided developing countries as a group are coal, oil, and
"See footnote 5.
10 A gigajoule (GJ), or 1 billion joules, is about& energy content of 8 gallons of gasoline. For reference, annual per-capita energy consumption in
the United States is 327 million British thermal units (Btu) or 343 gigajoules. An exajoule (EJ), or 10" joules, is about the same as a Quad (1.05 EI =
1 Quad).
u World Bank, World Development Report 1989 C™ Y ° rk > NY: Oxford University Press, 1989). Data, from p. 172, in tonnes Of oil equivalent (toe)
converted to gigajoules at 1 toe = 41.9 gigajoules.
12 Brazil, despite its relatively hi@ income, uses substantial quantities of biomass fuels in modern applications, such SS charcoal for steelmaking and
ethanol for cars. This contrasts with the use of biomass in the poorer countries, as a cooking fuel using traditional technologies.
13 The data in this section are taken from the World Energy Conference, G/oiwZ Energy Perspectives 2000-2020, 14th Congress, Montreal 1989 (Paris
1989); and the United Nations, 1986 Energy Statistics Yearbook (New York, NY: 1988), updated to 1987 by data provided by the UN. Secretariat.
8 • Energy in Developing Countries
Table 1-3-1985 Primary Energy Supplies (exajoules)
Primary Total Total
Coal ON Gas electricity commercial Biomass energy
World 88.7 104.6 58.2 33.0 284.5 36.9 321.3
Industrial countries 63.5 77.0 51.7 26.6 218.7 5.5 224.2
Developing countries 25.2 27.7 6.5 6.4 65.7 31.3 97.1
Share of industrial countries 72% 74% 89% 81% 7770 15% 70%
Share of developing countries 28% 26% 11% 19% 23% 85% 30740
NOTE: As in table 1-1, the values reported for developing country biomass are too low. Field surveys indicate that biomass accounts for roughly one-third of
the energy used in developing countries.
SOURCE: World Energy Conference, Global Energy Perspectives 2000-2020, 14th Congress, Montreal 1989 (Paris: 1989).
Figure 1-2 — 1985 Energy Consumption, Industrial and Developing Region Fuel Mix
Gas
7%
Primary^^iUliiiiili^'^^ Biomal
electricity y fuels
7% 32%
Developing nations
Total = 97 exajoules
Biomass
fuels
3%
Primary
electricity
12%
Gas
23%
Industrial nations
Total .224 exajoules
SOURCE: World Energy Conference, Global Energy Perspectives 2000-2020, 14th Congress, Montreal 1989 (Paris:
1989).
biomass (see figure 1-2). However, much of the coal
is used in India and China only. The other develop-
ing countries rely heavily on oil and biomass for
their energy supplies. Several developing countries-
China, India, Mexico, Brazil, and South Africa-are
among the world's 20 largest commercial energy
consumers (see table 1-4). China alone accounts for
almost 10 percent of the world's total commercial
energy use.
Three countries-China, India, and Brazil-
together account for about 45 percent of total
developing country consumption of both commer-
cial and biomass fuels. And these countries plus four
more-Indonesia, Mexico, Korea, and Venezuela-
account for 57 percent of the total. At the other end
of the scale are a large number of small countries
that, combined, account for only a small part of
global consumption. The 50 countries of Africa, for
example, use under 3 percent of total world commer-
cial energy consumption. Concerns about global
energy use and its implications focus attention on the
large consumers, but the energy needs of the small
developing nations, though of lesser importance to
global totals, are critical to their development
prospects.
The developing countries are becoming increas-
ingly important actors in global commercial energy.
Their share of the total has risen sharply in recent
years (see figure 1-3), from 17 percent of global
commercial energy in 1973 to over 23 percent now.
Despite their much lower levels of per-capita
commercial energy consumption, developing coun-
Chapter l-introduction and Overview %9
Table 1-4-Largest Energy Consumers, 1987
Per-capita
Total commercial commercial energy
energy consumption consumption
Country (exajoules) (gigajoules)
20 largest commercial energy consumers:
Rank
1 United States 68.1 280
2 U.S.S.R 54.7 194
3 China 23.5 22
4 Japan 13.4 110
5 West Germany 10.0 165
6 United Kingdom 8.5 150
7 Canada 7.5 291
8 India 6.5 8
9 France 6.1 109
10 Italy 6.0 105
11 Poland 5.3 141
12 Mexico 4.1 50
13 East Germany 3.8 231
14 Australia 3.2 201
15 Brazil 3.2 22
16 South Africa 3.2 83
17 Romania 3.1 136
18 Netherlands 3.1 213
19 Czechoslovakia 2.9 185
20 Spain 2.4 147
10 largest developing country energy consumers:
Rank
1 China 23.5 22
2 India 6.5 8
3 Mexico 4.1 50
4 Brazil 3.2 22
5 South Africa 3.2 83
6 South Korea 2.2 52
7 Argentina 1.7 56
8 Venezuela 1 .6 88
9 Indonesia 1.4 8
10 Egypt 1.0 20
NOTE: Data for the top 10 developing country energy consumers include
only countries listed in app. 1A.
SOURCE: United Nations secretariat.
tries accounted for one-half of the total increase in
global commercial energy consumption since 1973.
The increasing share of the developing countries
in global commercial energy consumption is widely
predicted to continue. The World Energy Confer-
ence projects an increase in their share to 40 percent
by 2020 (see figure 1-3), and this trend is confirmed
in a large number of other studies. 14 The developing
countries are projected to account for almost 60
percent of the global increase (over current levels)
in commercial energy consumption by 2020. China
alone accounts for over one-third of this increase.
These increasing shares are sufficiently large to have
a major impact on world energy markets. Despite the
more rapid rate of growth in energy consumption in
developing countries, their per-capita consumption
of commercial energy will still continue to be far
below the levels in industrial countries (see figure
1-4).
Analytic Focus
The analysis presented in this OTA study has
three important features. First, rather than concen-
trating on energy supplies, the analysis focuses on
the services energy provides. The reason for this
approach is simple. Energy is not used for its own
sake, but rather for the services it makes possible-
cooking, heating water, cooling a house, heating an
industrial boiler, transporting freight and people.
Further, there may be many different means of
providing a desired service, each with its own costs
and benefits. For example, transport is provided in a
number of ways-bicycle, motorcycle, car, bus,
light rail, or aircraft. The consumer chooses among
these according to such criteria as cost, comfort,
convenience, speed, and even aesthetics. Within
these consumer constraints, a more efficient car may
be preferable to increasing refinery capacity in order
to reduce capital and/or operating costs, or because
of environmental benefits. More than just engineer-
ing and economics must be considered, including
social, cultural, and institutional factors. Such fac-
tors are more readily included in a services frame-
work than in a conventional energy supply analysis.
Second, within this services framework the changes
in how energy is used are traced from traditional
rural areas to their modem urban counterparts. This
14 An analysis of projections of global commercial energy consumption over the next 20 years in Allan S . Manne and Leo Schrattenholzer, International
Energy Workshop: Overview of Poll Responses (Stanford University International Energy Project, July 1989), reports the results and assumptions of
over 100 projections of global energy consumption and production and provides the means of the different studies. Not all studies report results fall
regions. The coverage is nonetheless a comprehensive indicator of how energy forecasters view the futunThey suggest that the developing countries'
share could rise to over one-third by 2010. Longer term projections in general arrive at similar conclusions. For example, thEtllissions Scenarios
document, prepared by the Response Strategies Working Group of the Intergovernmental Panel on Climate Change, Appendix Report of the Expert
Group on Emissions Scenarios (RSWG Steering Committee, Task A), April 1990, concludes that, overa wide range of scenarios, the share of developing
countries (Centrally Planned Asia, Africa, Middle East, and South and East Asia) will increase from a 1985 reference level of 23 percent to between
40 and 60 percent of global energy in 2100, and that this group of developing countries would account for between 60 and 80 percent of the total increase
in energy consumption over this period. Further, developments in the developing countries define much of the difference between the low and high growth
10 • Energy in Developing Countries
Figure 1-3-Commercial Energy Consumption, 1973,
1985, and 2020 (developing nation energy demand as
a percentage of world total)
76%
60%
26%
0%
EZ Developing regions
Estimated
2020
SOURCE: World Energy. Conference. Global Enerqx Perspectives 2000-
2020, 14th Congress, Montreal 19S9 (Paris: 1989).
Figure 1-4-Per-Capita Commercial Energy
Consumption, 1973, 1985, and 2020
Gigajoulos per capita
200
150
100
50
- Developing regions
m Developed regions
1985
Year
2020
SOURCE: World Energy Conference, Global Energy Perspectives 2000-
2020, 14th Congress, Montreal 1989 (P&is: 1989).
lighting services can be met by using more conven-
tional lighting and increasing the amount of electric-
ity generated, by increasing the use of more efficient
light bulbs, or by a combination of the two. A
systems approach permits the comparison of effi-
ciency and supply options in achieving the desired
end.
In our analysis of energy services and systems it
is recognized that technology adoption and use is
embedded in an institutional framework that pro-
vides both incentives and disincentives to users, and
largely determines which and how technologies will
be used. This approach has a number of implications
both for the way technology is used now and for the
adoption of new technologies in the future. Thus, the
energy sector in many developing countries is
frequently characterized as "inefficient" in the
sense that more energy is used to provide a given
service or output than is usual in industrial countries.
In a wider context, however, taking into account the
many other relevant factors (financial, infrastructur-
al, managerial, and institutional), the technology
may well be used to the best of human ability and
often with considerable ingenuity and resourceful-
ness. I n many cases, although energy appears to be
used inefficiently, energy users may be acting
logically given the framework of incentives and
disincentives within which they make their deci-
sions. It follows therefore that the adoption of a new
technology will depend not only on the intrinsic
superiority of the technology itself but also on
whether institutional factors favor its adoption. The
policy environment is of crucial importance to the
adoption of new technologies.
progression from the traditional rural to the modern
urban helps illuminate the dynamics of energy use,
and how it can be expected to change in the future.
Third, the entire system needed to provide energy
services-from the energy resource to the final
energy service, including production, conversion,
and use-is examined as a whole. This is done in
order to show the total costs and consequences to
society, as well as to the individual, of providing
particular services, and how they might be provided
more effectively in terms of financial, environ-
mental, and other costs. For example, increased
Overview of the Report
Energy consumption in the developing world has
risen rapidly in the past and is widely expected to
continue increasing rapidly in the future. The World
Energy Conference, for example, projects (in its
"moderate" economic growth case) a tripling in
consumption of commercial energy in developing
countries between now and 2020. A survey of a large
number of projections of commercial energy use
broadly confirms this trend. The projected rate of
increase in commercial energy consumption implicit
in these forecasts is lower than that experienced
Chapter l--Introduction and Overview %11
between 1973 and 1985. 15 1 ncreased supplies of
biomass fuels (fuelwood and animal and crop
wastes) will also be required. The World Energy
Conference projects a 25 percent increase in biomass
use. Population growth and economic development
are the principal forces driving the rapid increase in
energy use.
Population Growth
In many developing countries, fertility rates (the
number of children expected to be born to a woman
during the course of her life) have dropped dramati-
cally over the past 20 years. Nevertheless, the
population of the developing world continues to
grow rapidly. Over 90 percent of world population
growth is now occurring in the LDCs. At present, the
population of the developing countries is about 4
billion, 77 percent of the world's population. Even
assuming continued decreases in fertility rates, the
population of these countries is projected to rise to
7 billion in 2025, 16 and could reach 10 billion in
2100, due to the large number of women of
childbearing age. Developing countries would then
account for 88 percent of the global population in
2100, and for virtually all of the increase in global
population. The increase in population alone in
developing countries would account for a 75 percent
increase in their commercial energy consumption by
2025 even if per-capita consumption remained at
current levels.
Economic Development
Securing higher living standards for the increas-
ing population of the developing world implies high
rates of economic growth. The World Energy
Conference, for example, assumes in its "moder-
ate" growth rate scenarios, an average annual gross
rate of economic expansion of 4.4 percent to 2020,
slightly lower than in the past. This would represent
more than a fourfold increase in economic activity
between now and 2020. Achieving such rates of
growth will certainly not be easy, especially in light
of the high levels of debt that have constrained
economic growth in the 1980's, and increasing
competition for foreign assistance from the countries
of Eastern Europe; but failure to achieve high rates
of growth could spell great hardship for the develop-
ing countries, as their populations are growing so
rapidly.
The process of economic development that under-
lies improving living standards in developing coun-
tries involves a number of changes, including higher
agricultural productivity, growth of manufacturing,
construction of a modern public works infrastruc-
ture, urbanization, and increased transportation (table
1-5 shows the rapid increase in the road transport
fleet). Higher standards of living also lead to
expansion in the ownership of consumer appliances
(table 1-6 illustrates saturation levels for some of the
most widely used appliances). All of these changes
have profound impacts on the amounts and types of
energy used.
Commercial energy consumption typically rises
faster than economic growth as the development
process gets underway, and the share of commercial
energy in total energy consumption grows as it takes
the place of traditional biomass fuels. Even though
the relative share of traditional fuels has declined,
the absolute amounts consumed have continued to
rise, by an estimated 2.5 percent per year. 17
Despite the strong connection between commer-
cial energy consumption and economic growth,
there is evidence of considerable differences among
developing nations in their energy intensity-the
amount of commercial energy consumed relative to
Gross National Product (GNP). There are examples
of countries with similar per-capita incomes that
consume quite different quantities of commercial
energy. Some of these differences result from
country-specific physical characteristics, but others
are associated with differences in social and eco-
nomic policies. Policies promoting heavy industry
and high rates of urbanization contribute to high
energy intensities. Similarly, the energy intensities
of countries change over time. In several industrial
countries energy intensities declined even before
1973 at a time when oil prices were falling, largely
,5 World Energy Conference, Global Energy Perspectives 2000-2020, Montreal, 1989 (Paris: 1989) projects a threefold increase in consumption of
commercial fuels in developing countries between 1985 and 2020. The Marine study projects a rise of 250 percent between 1985 and 2010 (see Alan
S. Marine and Leo Schrattenholzer, "International Energy Workshop: Overview of Poll Responses," Stanford University International Energy Project,
July 1989).
16 Rudolfo A. Bulatao, Eduard Bos > Patience \y. Stephens, and My x Vu, Europe, Middle East, and Africa (EMN) Region Population Projections,
1989-90 Edition (Washington, DC: World Bank, 1990), table 9.
"World Energy Conference, Global Energy Perspectives 2000-2020, 14th Congress, Montreal 1989 (Paris: 1989).
12 % Energy in Developing Countries
Table 1-5-Passenger Fleet Annual Growth in Selected Countries (percent)
Passenger Commercial Two and three
Country group cars vehicles wheelers Total
Developing countries:
Cameroon 11.8 29.5 9.1 13.1
Kenya 3.2 3.7 4.0 3.3
Bolivia 8.6 24.5 6.9 11.6
Brazil 8.9 7.3 25.6 9.8
Thailand 8.8 4.4 9.5 8.8
India 8.2 11.2 25.4 18.4
China 41.6 14.8 44.9 29.8
Taiwan 16.2 5.4 10.3 11.0
Weighted average 10.0 11.4 19.1 13.9
industrial countries:
Japan 3.0 4.1 7.0 4.4
United States 2.4 3.5 -5.6 2.3
West Germany 3.3 0.4 -2.2 2.6
Weighted average 2.6 3.6 2.4 2.8
SOURCE: Fleet size and growth from Energy and Environmental Analysis, "Policy Options for Improving Transportation Energy Efficiency in Developing
Countries," contractor report prepared for the Office of Technology Assessment, March 1990.
Table 1-6-Electric Appliance Ownership in Urban Areas (percent of households)
Country Radio Tv Fan Washer iron Cooker
China 39 66 45 2 1
Liberia ... , 76 4 56 1 74
Guatemala 78 25 — — — —
Manila ... , 80 78 82 — 94 16
Malaysia., 70 79 75 16 77 44
Hong Kong 90 91 96 34 87 91
Bangkok ., — 96 — 5 84
Taipei — 92 94 53 89
KEY: — information not available.
SOURCE: Jayant Sathaye and Stephen Meyers, "Energy Use in Cities of the Developing Countries," Annual Review of Energy, vol. 10, 1985, pp. 109-133,
table 6.
due to improved technologies. After the 1973 and
1979 oil price shocks, the decline in energy intensi-
ties was experienced in all industrial countries.
While the energy intensities of the developing
countries continued to rise after 1973, the rise was
less sharp than before. These experiences testify to
some flexibility in commercial energy use. The
current projections cited above, however, already
incorporate assumptions about declining energy
intensity in developing countries.
Energy Supply Constraints
The developing countries will face major difficul-
ties in tripling energy supplies over the next 30
years. Significant obstacles include financial con-
straints, difficulties in increasing biomass fuel
supplies, institutional and policy factors, and envi-
ronmental impacts.
Financial Constraints
Commercial energy consists of both domestically
produced and imported supplies. Many developing
countries rely on imported oil for virtually all of their
commercial energy needs. Further increases in
energy imports will impose a heavy burden on
limited foreign exchange resources, which may
already be under pressure because of debt service
payments (see table 1-7).
Funding the development of domestic energy
supplies and infrastructure also poses problems.
Energy supply facilities such as electricity generat-
ing stations and petroleum refineries are highly
capital intensive, placing major demands on the
scarce supplies of both domestic and foreign re-
sources available for capital investment. Already,
investments in the commercial energy supply sector
(including electricity, oil and gas, and coal) repre-
sented in the 1980s over 30 percent of public
Chapter l~Introduction and Overview »13
Table 1-7 — Energy Imports, Debt Service, and Export Earnings for Selected Developing Countries," 1987
Energy imports as share of Debt service as share of Energy imports and debt service
merchandise exports exports of goods and services as share of total exports 1 987
Country (percent) (percent) (percent)
Low-income:
Ethiopia 55 28.4 83.4
Tanzania 56 18.5 74.5
Madagascar 36 35.3 71.3
Rwanda 53 11.3 64.3
Benin 97 15.9 112.9
Kenya 39 28.8 67.8
Pakistan 26 25.9 51 .9
Burma 5 59.3 64.3
Lower middle-income:
Morocco 27 29.9 56.9
Jamaica 31 26.6 57.6
Turkey 31 31.7 62.7
Jordan 53 21.8 74.8
Syrian Arab Republic 40 16.5 56.5
Upper middle-income:
Argentina 10 45.3 55.3
Algeria 2 49^0 5U)
a Includes all nations in which debt service and oil imports combmedis greater than 50 percent ot total exports!
SOURCE: World Bank, Worid Development Report »9«9(New Ye *, NY: Oxford University Press, 1989), pp. 172-173 and210-21 1.
investment budgets in a wide range of developing
countries (see table 1-8), with the electric utility
sector accounting for the lion's share. Despite these
already large claims on capital resources, the current
level of investment in the electricity sector maybe
inadequate. The World Bank estimates that invest-
ments of $125 billion annually (twice the current
level) will be needed in developing countries to
provide adequate supplies of electricity .18 This
figure represents virtually the entire annual in-
crease in the combined GNP of the developing
countries.
Finding the domestic and foreign resources needed
to finance energy facilities has always posed major
difficulties for developing countries. Current levels
of domestic resource mobilization, often related to
low levels of energy prices, are reported in many
countries to fall short of the amounts needed for
system expansion. 19Inthe P ast > about one-half of an
investments in energy supply have been provided by
foreign sources, 20 but high levels of current debt in
many developing countries and increasing fiscal
difficulties in the industrial countries make it
difficult for many developing countries to increase
their borrowing from abroad.
Biomass Supply Constraints
It maybe equally difficult to increase supplies of
traditional biomass fuels. Despite rapid rates of
urbanization in the developing world, almost two-
thirds of the total populations in poor nations live in
rural areas. These populations largely depend on
biomass fuels to produce their energy, with some
rural electrification where available, and small but
vital quantities of petroleum products (for irrigation,
lighting, and transport).
Demand for biomass fuels (largely fuelwood) will
continue rising to meet the domestic needs of the
urban and rural poor, rural industry, and in some
cases, such as Brazil, modern industry. Overuse of
biomass resources already contributes to environ-
mental degradation (see below). Moreover, gather-
ing traditional supplies of fuelwood is time-
consuming, exhausting work frequently undertaken
by women and children, who are thus diverted from
other activities (education and farming) that could
18 U.S. Agency for Into national Development, P, mr Shortages in Developing Countries: Magnitude, Impacts, Solutions, and the Role of the private
Sector (Washington.L DC: March 1988), p. 10.
19 Lawrence J. HilL Energy Price Reform "> Developing Countries: Issues and Options (Oak Ridge, TN: Oak Ridge National Laboratory, August
1987).
^•WorldBank, The Energy Transition in Developing Countries (Washington DC: 1983).
14 % Energy in Developing Countries
Table 1-8-Estimated Annual Energy Investment as a Percentage of Annual Total
Public Investment During the Early 1980s
Over 40 percent
30-40 percent
20-30 percent
10-20 percent
0-10 percent
Argentina
Ecuador
Botswana
Benin
Ethiopia
Brazil
India
China
Egypt
Colombia
Pakistan
Costa Rica
Ghana
Korea
Philippines
Liberia
Jamaica
Mexico
Turkey
Nepal
Morocco
Nigeria
Sudan
SOURCE: Mohan Munasinghe, Electric Power Economics (London: Butterworths, 1990), p. 5.
eventually improve their productivity and living
conditions. An estimated one-third of the population
of developing countries now faces fuel wood defi-
cits, and will increasingly rely on crop wastes and
animal dung to meet their energy needs.
Institutional Constraints on Rapid Expansion
in Energy Supplies
Over and above the capital constraints discussed
above, a wide range of other factors, customarily
defined as "institutional," currently impede com-
mercial energy sector development. While defini-
tions of institutional factors differ between observ-
ers, they are generally taken to be nontechnological,
encompassing a variety of economic, organizational,
and policy factors that affect the way technologies
perform in operational settings. 21 Some of these
factors (the worldwide increase in interest rates, for
example) are outside the control of individual
countries, but others are related to policies and
procedures in the individual country.
The electricity supply system offers an example of
the importance of institutional factors. In most
countries of the developing world the electricity
sector is government owned, reflecting the impor-
tance attached to electric power for meeting eco-
nomic and social objectives, and in some cases,
especially in small systems, the advantage of cen-
tralization for securing economies of scale and
coordination in planning and operations. However,
government ownership can lead to interference and
loss of autonomy in day-to-day management of
utility operations and therefore reduced efficiency.
A recent World Bank report on the power sector in
developing countries points out that:
Such interference has adversely affected least cost
procurement and investment decisions, hampered
attempts to raise prices to efficient levels, mandated
low salaries tied to civil service levels, and promoted
excessive staffing. This in turn has resulted in
inadequate management, the loss of experienced
staff due to uncompetitive employment conditions
and poor job satisfaction, weak planning and de-
mand forecasting, inefficient operation and mainte-
nance, high losses, and poor financial monitoring,
controls and revenue collection. 22
Manpower problems are exacerbated by the lack
of standardization of equipment, 23 which makes the
learning process more difficult. Another disadvan-
tage of the multiplicity of equipment is the difficulty
of maintaining adequate supplies of spare parts.
Pricing policies are frequently identified as a
major institutional problem. Energy pricing policies
vary in developing countries, reflecting differences
in energy resource endowments and social and
developmental policies. However, price controls on
energy products, such as kerosene and some electric-
ity prices, are a common feature in many countries.
While low prices help to make energy more afforda-
ble, they can also result in a level of revenues
inadequate to cover costs and finance future supply
expansion. Many analysts have characterized this as
a common problem in the electricity sectors of a
wide range of developing countries. 24
21 World Bank,/! Review of World Bank Lending for Electric Power (Washington, DC: Match 1988), p- 74.
^Mohan Munasinghe, Current Power Sector issues in Developing Countries (Washington, DC: World Bank, November 1986), p. 14.
Bin Mali f° r example, there ar e 40 diesel generator sets fro. 17 different manufacturers (see U.S. Agency for International Development, "Electric
Power Utility Efficiency Improvement Study," draft core report, May 15, 1990).
mhui, op. cit, footnote 19 ? aaA Donald He rtzmark, ' 'Energy Efficiency and Energy Pricing in Developing Countries," OTA contractor report, June
1990.
Chapter l~Introduction and Overview 15
Figure 1-5-Suspended Particulate Matter Levels in Selected Cities, 1980-84
Shown is the range of annual values at individual sites and the composite 5-year average for the city.
Key
1. Kuwait
2. Shenyang
3. Xian
4. New Delhi
5. Beijing
6. Calcutta
7. Tehran
8. Jakarta
9. Shanghai
10. Guangzhou
11. Illigan City
12. Bangkok
13. Bombay
14. Kuala Lumpur
15. Zagreb
16. Rio de Janeiro
17. Bucharest
18. Accra
19. Lisbon
20. Manila
10
WHO Guideline 60-90 ug/m 3
1
1,000
I
Range of individual site
annual averages
Combined site
average 1980-84
TT
I I
"i — r
TT"
]."
JJJ (
1
1 15.
•I 16 -
| 17.
18.
/
19.
20.
SOURCE: World Health Organization and United Nations Environment Fund, Global Pollution and Hearth (London: Yale University Press, 1987), figure 3.
Environmental Degradation in Developing
Countries
Developing countries are experiencing accelerat-
ing rates of environmental degradation and pollu-
tion. While many factors contribute to environ-
mental degradation, energy production and use play
key roles, especially in urban environmental quality.
Even at present levels of energy generation and use,
the impacts on environmental quality are severe in
many areas. Additional large increases in energy use
will exacerbate the situation unless steps are taken to
mitigate adverse environmental impacts. At the
same time, energy is an essential input to such
environmental control systems as sewage treatment.
The combustion of fossil fuels has led to levels of
air pollution in cities of developing countries that are
among the highest in the world (see figures 1-5 and
1-6). The transportation sector is the largest contrib-
utor to air pollution in many cities. 25 The combus-
tion of oil or gas in stationary sources, such as
electric generating units, factories, and households,
also contributes through emissions of nitrogen
oxides, particulate, sulfur dioxide, carbon dioxide,
carbon monoxide, and hydrocarbons. The fossil fuel
mix has an important impact on emission levels.
Coal is the most deleterious of fossil fuels in terms
of emissions per unit of useful energy provided,
particularly when it is not burned in modern,
well -operated plants.
In addition to the environmental damage caused
by the combustion of fossil fuels, their production
and transportation also impose environmental costs,
such as disturbance of lands and aquifers from coal
mining, and accidental leaks and spills during oil
and gas production and refining. Air quality impacts
include, for example, local air pollution from
particulate and other emissions during coal mining
preparation, and transport and the release of meth-
ane, during coal mining and natural gas production
and transportation.
Non-fossil energy sources such as hydroelectric
development also causes environmental damage.
Dam construction often requires the clearing of
lands for access routes and removal of construction
material, with resulting soil degradation and erosion.
Filling the reservoir floods large tracts of land,
which usually means loss of agricultural land,
human settlements, fish production, forests, wildlife
^In Indian cities, for example, gasoline-fueled vehicles — mostly two and three wheelers — ar e responsible for 85 percent of carbon monoxide and 35
to 65 percent of hydrocarbons, while dieselvehicles — buses and trucks-are responsible for over 90 percent noNO x emissions. Tats Energy Research
Institute, TERI Energy Data Directory and Yearbook 1988 (New Delhi, India: 1989), p. 250.
16 • Energy in Developing Countries
Figure 1-6-Sulfur Dioxide Levels in Selected Cities, 1980-84
Shown is the range of annual values at individual sites and the composite 5-year average for the city.
1
Concentration (ug/m 3 )
10
WHO Guideline 40-60 ug/m 3
/ \ 100
I
Key
I
Range of individual site
annual averages
I I
I I I I I
i i i i i
I I
1 1 1
.
2.Shenyang
I
1
I 2-
L_
I | 3
1 1 Is.
4. Seoul
I I I
5. Rio de Janeiro
4
Combined site
average 1980-84
6. Sao Paulo
7. Xian
8. Paris
i c
J — 1.1
I 18.
9. Beijing
10. Madrid
11. Manila
12. Guangzhou
13. Glasgow
14. Frankfurt
i
i —
l
1 p.
— r — m —
i in.
-12.
J 113.
| 14.
15. Zagreb
i
1 5.
16. Santiago
17. Brussels
1
I 18.
]17.
18. Calcutta
19. London
20. New York City
i i
i
i
116.
119.
[20.
SOURCES: World Health Organization and United Nations Environment Fund, Global Pollution and Health (London: Yale University Press, 1987), figure 2.
habitat, and species diversity. 26 The dam and reser-
voir interrupt the free flow of surface water, affect
water tables and groundwater flow, and disrupt
downstream flows of water and nutrient-laden sedi-
ments. These environmental costs are all the more
onerous if the lifespan of hydroelectric projects is
reduced through heavier than expected siltation of
reservoirs from deforested and/or degraded lands
upstream. 27
Burning biomass, the source of energy used by
most of the developing world's population, also
causes environmental degradation. Although the use
of biomass for fuel is only one, and not the principal,
cause of deforestation, it does add additional pres-
sure on forest resources especially in arid or
semi-arid regions where forest growth is slow and
where there is a high population density or a
concentrated urban demand for fuelwood, such as
the African Sahel. 28
When fuelwood is in short supply, rural popula-
tions turn to crop residues and dung for their fuel
needs. To the extent that these forms of biomass
would have been used as fertilizers, their diversion
to fuel contributes to lowered soil productivity. 29
Finally, biomass fuel combustion has a significant
impact on air quality. Food is typically cooked over
open fries or poorly vented stoves, exposing house-
hold members-particularly women and children —
to high levels of toxic smoke. Similarly, in colder
climates, homes in rural areas are often heated by
open fires, with increased exposures to toxic smoke. 30
^For example, the reservoir Akosombo on the Volta in Ghana, with a land requirement of 8,730knr, approaches the size of countries such as Lebanon
or Cyprus. See R.S. Panday (cd.), Afan-marfe Lakes and Human Health (Paramaribo: University of Suriname, 1979).
zijbs Hirakud reservoir in India, f° r example, was expected to be productive for about 110 years, but now has an estimated productive lifetime of
35 years. U.S. Congress, Office of Technology Assessment, Technologies To Sustain Tropical Forests, 0TA-F-214 (Washington, DC: U.S. Government
Printing Office, March 1984), p. 43.
^Douglas F. Barnes, World Bank, "Population Growth, Wood Fuels, and Resource Problems in Sub-Saharan Africa," Industry and Energy
Department Working Paper No. 26, March 1990; RJP. Moss and WJB. Morgan, ' 'Fuelwood and Rural Energy Production and Supply in the Humid
Tropics," United Nations University, Tycooly International Publishing, Ltd., Dublin, 1981 ; Daniel Finn, "Land Use and Abuse in the East Africa
Region, "AMB/O, vol. 12, No. 6, 1983, pp. 296-301, Dennis Anderson and Robert Fishwick, World Bank, "Fuelwood Consumption and Deforestation
in African Countries," staff working paper No. 704, 19S4.
^Organic matter in soils provides most of the nitrogen and sulfur and as much as half the phosphorus neededbyplants.lt helps the soil bind important
minerals such as magnesium, calcium, and potassium that would otherwise be leached away. Itbuffers the acidity of soils, and it improves water retention
and other physical characteristics. See Geoffrey Barnard and Lars Kristoferson, Agricultural Residues as Fuel in the Third World, Earthscan,
International Institute lor Environment and Development, Energy Information Program, technical report No. 4 (Washington DC and London: Earthscan,
1985).
30 Kirk R. Smith, Biofitels.Air Pollution, and Health: A Global Review (New York, NY: Plenum Press, 1987).
Chapter 1 --Introduction and Overview »17
Greenhouse Gases and Developing Countries
Energy use in developing countries also contrib-
utes to increased emissions of greenhouse gases and
associated global climate change. An international
panel of scientific experts of the Intergovernmental
Panel on Climate Change (I PCC) 31 recently con-
cluded that:
. . . emissions resulting from human activities are
substantially increasing the atmospheric concentra-
tions of the greenhouse gases: carbon dioxide,
methane, chlorofluorocarbons (CFCs) and nitrous
oxide. These increases will enhance the natural
greenhouse effect, resulting on average in an addi-
tional warming of the Earth's surface. 32
Based on current models, the panel predicts that,
under a "business as usual" scenario, the global
mean temperature will increase at a rate of about 0.3
°C per decade during the next century, a rate at least
10 times higher than any seen over the past 10,000
years. 33
Atmospheric concentrations of C0 2 have in-
creased by about 25 percent since preindustrial
times, largely due to emissions from the burning of
fossil fuels and from deforestation. In 1985, accord-
ing to another I PCC working group, developing
countries contributed about one-quarter (26 percent)
of annual global energy sector Commissions. %
Under the "business as usual" scenario, with
expanding populations, rapidly increasing energy
use, and assuming the absence of control measures,
the developing country share would increase to 44
percent of annual energy sector emissions by 2025. 35
The magnitude of C0 2 emissions from fossil fuel
sources is fairly well known, but the contribution
from deforestation, which is virtually all from
developing countries, cannot be estimated accu-
rately. This makes it difficult to calculate with
confidence the developing country share of global
annual and cumulative emissions for C0 2 and other
gases. Estimates of C0 2 emissions from tropical
deforestation differ by a factor of 4. 36 By various
estimates, deforestation could be the source of
between roughly 7 and 35 percent of total annual
C0 2 emissions.
Despite uncertainties, it is safe to conclude that
the developing countries already contribute a sub-
stantial part of current annual global C0 2 emissions,
and that their share will increase in the future. But,
because of their large and rapidly growing popula-
tions, per-capita Commissions in the developing
countries will still remain much lower than in the
countries of the industrial world. Developing coun-
tries also account for at least half of the global
anthropogenic generation of two other important
greenhouse gases, methane and nitrous oxide.
Developing and industrial countries would both
be damaged by the anticipated impacts of climate
change. In addition to increases in mean global
temperature, other major effects of global climate
change would include increases in sea level 37 and
shifts in regional temperature, wind, rainfall, and
31 The Intergovernmental Panel on Climate Change (IPCC) is an international task force created by the World Meteorological Organization and the
United Nations Environment Program. Initiated in 1987, this body focuses on analyzing current information available on climate change issues and
devising strategies to address climate change.
32 Intergovernmental Panel on Climate Change, "Policymakers Summary f the Scientific Assessment of Climate Change: Report to IPCC from
Working Group I," June 1990, p. i.
33 Id., at p. ii. This would mean an increase over the preindustrial global average temperature of 2 degrees centigrade by 2025and 4 ' C by 2090. This
best estimate prediction has an uncertainty range of 0.2 to 0.5'C per decade. This was based on projections derived from another working group that
emissions of C0 2 could grow from approximately 7 billion tonnes of carboiflBtC) in 1985 to between 11 and 15BtC in 2025. Methane emissions were
projected to increase from 300 teragrams in 1985 to 500 teragrams by 2025.
^Intergovernmental Panel on Climate Change, Working Group m (Response Strategies Working Group), "Policymakers' Summary Of the
Formulation of Response Strategies: Report Prepared foiIPCC," Executive Summary,p. 10, table 2, June 1990.
35 OTA calculation based on 2030 ~@ Emissions — Lower Growth Scenario, tables A-21 and A-193 in Intergovernmental Panel on Climate Change,
"Emissions Scenarios Prepared by the Response Strategies Working Group of the Intergovernmental Panel on Climate Change," Appendix Report of
the Expert Group on Emissions Scenarios (RSWG Steering Committee, Task A), April 1990.
^IPCC, "Policymaker's Summary of the Formulation of Response Strategies: Report Prepared for JPCC by Working Group HI, " June 1990, p. 5.
Estimates of C0 2 emissions from deforestation and land use changes (including wooduel) were 0.6 to 2.5 PtC in 1980 according to a report prepared
for the IPCC Working Group 1. "Scientific Assessment of Climate Change: Peer Reviewed Assessment for WG1 Plenary Meeting, May 1990," Apr.
30, 1990, p. 1-9. More recent evidence from selected countries would indicate deforestation rates now are higher than they were in 1980. Estimates of
C0 2 emissions in 1987 were 5.7 +0.5 PtC from fossil fuel burning and total emissions were of 7 +l.lPtC.
37 The IPCC working group predicted an average rate of global mean sea level rise of about 6 cm per decade over the next century-20 cm by 2030
and 65 cm by the end of the century (wittsignificant regional variations). This predicted increase is due primarily to thermal expansion of the oceans
and melting of some land ice.
18 • Energy in Developing Countries
storm patterns. These effects in turn would sub-
merge some low-lying coastal areas and wetlands,
threaten buildings and other structures in these areas,
and increase the salinity of coastal aquifers and
estuaries. Such changes would disrupt human com-
munities and aquatic and terrestrial ecosystems, and
affect food production and water availability. A
number of developing countries would be especially
vulnerable to rising sea levels through threats to
coastal communities and ecosystems, increased
vulnerability to storm damage, and inundation of
low-lying arable lands. 38 The adverse effects of
climate change would exacerbate the impacts of
increased populations in Asia, Africa, and small
island nations of the Caribbean, Indian, and Pacific
Oceans. 39 The timing, severity, and extent of these
potential impacts remain uncertain.
Prospects for Efficiency
Improvements in Energy
Production and Use
Increasing energy supplies in the "business as
usual' mode that underlies the conventional projec-
tions thus poses formidable challenges. This unfor-
tunate fact suggests that alternative approaches for
providing the vital energy services needed for rapid
economic and social development should be investi-
gated, focusing on efficiency improvements.
OTA examined the following energy-related serv-
ices: cooking, lighting, and appliances in the resi-
dential and commercial sectors; process heat and
electrical and mechanical drive in industry and
agriculture; and transportation (see ch. 3). For the
developing countries as a whole, the largest energy
end use markets are residential/commercial and
industry, which together account for roughly 85
percent of the energy used by final consumers,
including traditional energy. Considerable differ-
ences exist, however, among developing nations. In
Africa, the residential/commercial sector constitutes
a particularly high share (mostly in the form of
biomass fuels for cooking), while industry's share is
quite low. In Latin America, transportation accounts
for an exceptionally high share of the total, whereas
its share in India and China is quite low.
Industrial process heat and cooking are the largest
energy services, each accounting for about one third
of all energy consumed (commercial plus biomass)
in developing countries. This pattern of energy use
contrasts with the United States, where transporta-
tion and space conditioning are the highest (although
in the United States process heat is a major user as
well). Much of the energy used for residential
cooking and process heat in the developing countries
is consumed in China and India. Together their
consumption of energy for cooking and industrial
process heat accounts for over 40 percent of all
cooking and process heat energy used by developing
countries, and for well over one-quarter of all the
energy consumed in developing countries.
A wide range of technologies are currently used to
provide energy services in developing countries. For
example, cooking technologies include stoves using
fuelwood, charcoal, kerosene, liquid petroleum gas,
natural gas, and electricity, all with different charac-
teristics. These technologies vary widely in their
energy efficiency. In an open fire, for example, only
about 15 percent of the energy contained in fuel-
wood goes into cooking. In contrast, in a' 'modern"
gas stove about 60 percent of the energy contained
in the gas is used in cooking. The wide range of
efficiencies in the current stock of stoves suggests
opportunities for increasing efficiencies of the stock
and therefore providing more cooking services with
less energy.
There are also differences in efficiencies in
providing energy services in the industrial sector-
industrial process heat and electric and mechanical
drive. The two largest developing country energy
consumers, India and China, currently rely on
several technologies that are a generation or more
behind the state of the art, and are much less
energy-efficient than technologies now being used
in the United States and other countries. Integrated
iron and steel plants in China and India, for example,
use twice as much energy per ton of crude steel
produced as integrated plants in the United States
and J apan. Lower efficiencies are also frequently
observed in the transportation sector.
An analysis of the energy supply industry in
developing countries similarly indicates much lower
^Intergovernmental Panel on Climate Change, "Policymalcers'Summary of the Potential Impacts of Climate Change: Report from Working Group
H to the IPCC," May 1990, p. 8.
39 J,D. Milliman e * al., "Environmental and Economic Implications of Rising Sea Level and Subsiding Deltas: The Nile and Bengal Examples,"
AMBIO, vol. 18, pp. 340-345, 1989.
Chapter 1 -Introduction and Overview »19
operating efficiencies than in the industrial coun-
tries. In electricity generation, for example, thermal
power plants frequently operate far below design
capacity and efficiency. Transmission and distribu-
tion losses (including unaccounted for losses, un-
metered use, etc.) are frequently over 15 percent,
substantially higher than losses in industrial country
systems. 40 Refineries a/so operate at much lower
efficiencies.
Energy supplies in many developing countries are
unreliable, imposing a heavy economic burden. In
India, for example, losses sustained by industry due
to unreliable electric power supplies in recent years
are estimated to represent 2 percent of annual GNP,
not including losses in agriculture or losses and
inconvenience experienced by residential and com-
mercial users. Similar losses have been estimated for
Pakistan. Furthermore, electricity supplies in many
countries are of poor quality, discouraging the use of
efficient technologies that are critically dependent
on high-quality energy supplies.
In characterizing important parts of the energy
system as "inefficient, however, it should be
realized that in many cases users and producers are
acting logically given the framework of resources,
incentives, and disincentives within which they
make their decisions. One of the reasons that poor
households use fuel wood inefficiently is that they
lack the financial means to buy more efficient
cooking systems. Industrial users must cope with
antiquated machinery and erratic fuel supplies of
uncertain quality. On the supply side, the record of
"poor" performance reflects many factors: poor
repair and maintenance, unavailability of spare
parts, low fuel quality, older equipment, unsatisfac-
tory management, lack of skilled workers, problems
of reaching dispersed populations served by inade-
quate transport systems, and inappropriate pricing
and allocation systems.
The existence of wide differences between opera-
tional efficiencies in reasonably standardized opera-
tions (e.g., cooking, steelmaking, electricity genera-
tion, and petroleum refining), both among develop-
ing countries and between the developing and
industrial countries, suggests that dramatic improve-
ments in efficiencies are possible. However, the
importance of factors other than technology must be
recognized for the role they play in improving
efficiencies. The policy environment in particular is
crucial to the adoption of new technologies.
More efficient ways of providing energy services
for development, including both technologies and
the institutional and policy mechanisms determining
their rate of adoption, will be presented in a later
report of this OTA assessment. Attention will also be
paid to the energy implications of different develop-
ment strategies. Some development strategies are
associated with high energy use. But developing
countries at the beginning of the development
process may be able to capitalize on technology to
develop toward modern economies without the high
energy growth that earlier characterized the path to
industrialization.
'Win the United States, for example, transmission and distribution losses in dense urban service areas are between 6 and 7 percent and in rural service
areas nearer 9 to 10 percent.
Appendix 1 A— Economic, Social and Energy Indicators for Developing Countries
0.
z
'5.
2. 8.
• ss
52
ll
Is
o Q-
<5 «
Mi.
£"
1 1
"5
r
$ "ST
a a:
~ 2
I§
CO o
^ 3
J?
2 ■*
8-00
3-2
t? 1
-EiS a.!Z
"5 £~
is
3 Q.
» M
O <0
05
10
1
1|
§8
82
o «
is.
j2 m
O Q>
11
a.
12
s£-
•sr:
a to
13
lis
• o E
lis
14
CD
"I
« 8.
8-1
» = oo
1-S
I to
It?
Low-Income countries
Average 280 w —
1 Ethiopia 130 304
2 Bhutan 150 —
3 Chad 150 254
4 Zaire 150 200
5 Bangladesh 160 646
6 Malawi 160 —
7 Nepal 160 526
8 Lao PDR 170 —
9 Mozambique 170 528
10 Tanzania 180 291
1 1 Burkina Faso 190 359
12 Madagascar 210 370
13 Mali 210 337
14 Burundi 250 338
15 Zambia 250 579
16 Niger 260 320
17 Uganda 260 319
18 Somalia 290 348
19 Togo 290 489
20 Rwanda 300 388
21 Sierra Leone 300 441
22 Benin 310 531
23 Central African Rep. . 330 436
24 Kenya 330 603
25 Sudan 330 562
26 Pakistan 350 1,165
27 Haiti 360 591
28 Lesotho 370 795
29 Nigeria 370 565
30 Ghana 390 361
31 Sri Lanka 400 1,491
32 Yemen, PDR 420 1,038
33 Mauritania 440 539
34 Indonesia 450 1 ,269
35 Liberia 450 539
37 Burma — 562
38 Guinea — —
39 Kampuchea, Dem — —
40 Viet Nam — —
China and India
Average 300 w —
41 China 290 1,489
42 India 300 775
^w
42
51
43
32
47
37
57
50
61
38
43
54
59
12
34
76
65
29
37
45
46
41
31
37
23
21
30
51
27
16
37
26
37
30w
31
30
24 W
12
5
30
38
13
13
9
17
23
29
8
23
19
7
53
18
10
36
24
7
26
39
45
22
21
31
29
19
33
32
21
42
38
27
42
24
24
33 w
38
27
54 w
47
48
46
52
51
46
51
48
48
53
47
54
47
49
53
45
48
47
53
49
41
50
50
58
50
55
55
56
51
54
70
51
46
60
54
60
42
42
66
S3w
69
58
103w
154
128
132
98
119
150
128
110
141
106
138
120
169
112
80
135
103
132
94
122
151
116
132
72
108
109
117
100
105
90
33
120
127
71
87
70
147
46
62 w
32
99
13,550 w
77,360
23,310
38,360
6,730
11,560
32,710
1,360
37950
57,180
10,000
25,390
21,120
7,100
38,770
21,900
16,090
8,720
34,680
13,630
15,940
23,070
10,100
10,110
2,900
7,180
18,610
7,980
14,890
5,520
4,340
12,110
9,460
9,240
3,740
57,390
1,000
1,640 w
1,000
2,520
9w
12
4
6
18
4
25
19
7
3
6
36
7
4
19
6
12
21
3
16
13
20
20
18
18
22
35
66
15
41
43
25 w
42
35
8.19
22.31
5.85
8.68
2.46
7.97
8.64
10.00
9.59
9.29
7.83
5.87
6.03
7.20
13.19
5.44
7.01
7.72
1.88
8.59
20.00
9.77
10.74
14.43
8.31
2.19
9.51
0.00
8.32
6.10
4.76
0.00
0.00
7.70
17.39
4.07
5.69
3.40
1.70
3.01
0.67
0.00
0.57
3.01
0.03
1.77
0.85
3.42
0.89
156
0.72
1.19
0.64
0.60
19.58
1.47
1.15
2.63
1.56
1.25
1.84
1.40
0.74
2.94
2.12
8.49
1.80
0.00
4.82
5.07
4.82
15.22
4.21
8.47
4.78
2.65
2.46
3.52
21.15
8.34
8.86
22.31
6.42
11.69
2.49
9.75
9.49
13.42
10.48
10.54
8.55
7.06
6.67
7.80
32.78
6.91
8.15
10.35
3.44
9.84
21.84
11.16
11.48
17.38
10.43
10.67
11.31
0.00
13.14
11.18
9.57
15.22
4.21
16.17
22.17
6.72
8.15
6.92
22.00
11.36
92
100
91
74
99
82
91
75
92
88
92
83
90
92
40
79
86
75
55
87
92
88
94
83
80
20
84
63
55
50
48
78
61
70
87
49
7
27
100
100
100
X
40
100
92
75
100
100
100
100
100
100
33
100
100
100
100
100
100
x
100
100
100
39
93
x
100
87
70
100
x
100
1
100
100
20
x
14
16 w
55
2
21
10
31
56
7
36
32
8
11
9
17
9
8
53
10
97
1
39
38
26
16
310.0
14
25
8
13
11
5
5w
2
17
21 .9 W
28.4
3.9
12.8
24.2
23.3
9.7
18.5
35.3
9.9
38.5
13.5
33.5
19.5
8.3
14.2
11.3
15.9
12.1
28.8
6.8
25.9
7.0
4.4
19.2
19.2
38.2
18.2
27.8
2.5
59.3
10.7
7.1
18.9
Chapter 1— Introduction and Overview %21
I-- ■* r- CM CM CO 0> U) O I
CMCMCMCMCMCMCM*-i-
oi CO
1 maioiai-viri-^dmai-^mreiai-^cito-id
f-T-T-CMCM^-CMCM « WP) OIt- CM *-*-»- CO
~- CM CM ■* ■* cm eg
3 ■* CM SO t- I l~-UJOI--»-lOC2lO-*lO'i-<0*-OCO<010"-eMO>»-f~COCMg'<t'-0
CM
> o cm co -- o ■*
O X CO CO o o
O CMOINO
OOMNI
o o o> o> I
t ■* X O O CO X o
to O O h* o
SS I 2 X S ^
ICMCOr^-Q ICO-^COmcOQCnLncDQCOCMCO-^CO I Ifl nN CM OC4 O O) o n I
I «li-«5 I CDCMIOCO-Sf^fin* <0-3FincMI>-CM»-'-'-in \F *-
> CO O CM O 0> IO CO •
CMCM-^-^^-^-CM-r-CMCO — -^CMCM
I incocMin^-cocjwocMO
S'coiri^Kiodr^uicMco
^COCOCOIOCOOCMCM-^
I NONi-O<0000)C0(DnoU)SQQU)i-<>)( , >O'tNr-<D*-Ot-e000(0inO I
I ioo>iD«on«'-iOMMOo>>oifiioo>iiioinoNO>n<-<MO<MocaN4N I
icqqincnqcq^oqoqt^cncM
irjwsibbbbui^Ndoib^o&woicarJtfidriNtfi^^ffldT-'d^N^
cocor^i-^^-r^cMi
sdo^d^ni^w^b
t- t- CM
> CO i- <o <o CO io
S CO O CM I© CD <
CO*- i- t- CD CM CO CM CM *-
IOW--T-r^i-i- i- i- j-
SC0C0C0C0^^^^C0CM^C0^CMC»*CMC0^inC»lQ^C0CM-t-h-C0^tinO01Q
I 1
> IT) *- CM -r- t
* ©NNM
I $M5t , ~S9 : i *r:S ,v *Sfi^r: a>, * CJ<D I ;? r^ <o co co i^ © | t-wo>con
IC\lCM*-OICM^-CMCO'r-TONCM'r-i-i- I Ol OJ *- *-■»-▼- I *- ■»- i-CJ
I i£ n i$ fn I JQ^WT-iflNWNNnNOaT-NP) CMCOCO^OlCM^^r^CD
h* i-^CJ> CO Wo5s0)5S8n5nSi-$S*O SScQl^f-^SwSS
■^ ^ 7-" t-" t-" T-" r-W" W T-* T-* *-"* of -r-* CM" Of CO Cvf *-" CM 04 OI C0~ CO CO
*88888°888°°
^ — iniowmoocor^h-r-co
> a> o> en o> o o ^ cm cm co_""t-*iq to ©co co co *°oi-cvi(0<o«o.-^., - —
"^?CcMorcMorc\roforco*'*ioio
o o o o o
o co *- <o t-
r*- 01 01 ^ co
1111
JNMS
(0 0)0
CO CM t-
*- T- CO
888
co in eg
o°. «
o ■<»
*2E,$8!238 l$8838Si I I 8S8BS858RSK I 3SS8 I *8sSSSScm§S I S
iqooooooo looooqoo ioooooooqoooooq i 500000000000 *
jgoo
•" co to r-
g CO CO •»
5 00 co c
iiomit
jinujTtlflSGDcomivivo
'h.coincO'3-co-vcococo*-
|»-eoo>coeM*-»-co-«|m 1
1
CC • CO -„
!<3<
SQ*-CMC0'*IQ(C>l s *.C00>O'«-CMC0'
cocococococococococDTs.r^.r-.rvi
■1 Si Sill s.S
a<<^0?i=ZiO
cocoeoctS353cnO)Oi
Him
0)90(00
I
a.
a.
a ■
■= •
o o>
c co
j& «
o * o> :
_ " / 1
S x 1
o
T
Chapter 2
Energy and Economic Development
Contents
Page
Introduction and Summary ... .*. * , ** *. **. ..**, *,** ***. * + . *****, * *e. 25
Economic Development and Its Impact on Energy .....**,.* **** f ***.** **** 25
Risino Pooulations **** + * * ** ********** ******p* _,.
Higher Living Standards .*. ..* **. ... ^***. **** .**. ..** *. oc *. 27
Changes in Energy Consumption With Economic Development c . .27
projected Energy Consumption in Developing Countries v *...****.*....„+„.**.**,* 33
The Energy Sector and the Macroeconomy ... .*. ..** * f . ... ,.. * *+ *****5 + ****'_* 34
Financing Energy Supplies * * *** ** *** * ** ** ** **** ** *q*** ~.
Energy Supply Reliability * * ** **_)_***** ** ** ** *****_)_** * * «.
Energy Pricing and Demand Management ,* * *, *+**£* ** i ** i q*. *+ 35
Energy and The Traditional Sector * * , , Jjp^ **, **. m ***** ^
Inequities in Resource Distribution and Access 42
The Role of Women ** * **** *** ** * **** ** * * *** ** .^
The Role of Commercial Biomass in the Rural Economy ** * "* **** 40
Conclusion * * * ** *^* **** **** ** ** *** * *** * **rvc& AA
Boxes
B0X Page
2-A. Factors Affecting Population Increase * ( -im i * i ** j ( #*+ 23
2-B. Energy Conservation Initiatives in ASEAN Countries „ >••>••> < ,
Figures
Fl S ure Paee
2-1. World Population Growth, 1750-2100 in Industrial and Developing Regions Zb
2-2. Average Annual Increase in Population Per Decade in Industrial
and Developing Regions, 1750-2100 ** *. .,. *.. ,*, * 26
2-3. Historical and Projected Global Population, 1950-2025 ( _ * #< 27
2-4. Projected Shares of Global Population, 2000 and 2025 ."'.',."..*....'..,+*,*,.**'.' 27
2-5. Commercial Energy Consumption and Economic Development in
Selected Countries * * * *** ** **** * ***** ** ** _.
2-6. Commercial Energy Consumption and Social Development
in Selected Countries **** ** * * * * *** ** *** ***** .._
■ ■ ' ■$' 1 •• 111 • 1 11 + + vc <a" ' J£
2-7. E nergy I ntensity and Economic Development, 196O-2000 33
2-8. Historical and Projected Energy Consumption in Developing Countries:
World Energy Conference "Moderate" Projections * ********* ^
2-9. Seasonal Pattern of Energy Expenditure on Agricultural and OtVier Essential 'Tasks:
Adult Farmers, Genieri Village , .*. .*, ,., * ** ii **.+**** i *+ ^
Tables
Table Page
2-1. Energy Intensities in Selected Countries ** +,**.,. ...@m=****.*.* ^2
2-2. Commercial Energy Consumption, 1985 and Projections for 2010 35
2-3. Estimated Annual Energy Investment as a Percentage of Annual Total Public
Investment During the Early 1980s , * ,,.*+..*,++*** 35
Chapter 2
Energy and Economic Development
Introduction and Summary
This chapter examines the two-way linkage be-
tween energy and economic development in devel-
oping countries-- how the process of economic
development impacts energy, and how, in turn,
developments in the energy sector can affect eco-
nomic growth.
In the course of economic development, commer-
cial energy consumption is observed to increase
faster than economic activity. There are a number of
reasons for this: the growth of mechanized agricul-
ture and manufacturing, the construction of a mod-
em infrastructure, urbanization, increased transpor-
tation of goods and services, rapid expansion in
ownership of consumer appliances, and the substitu-
tion of commercial for traditional fuels. The absolute
amount of traditional energy consumed also contin-
ues to rise, although its share of total energy
consumption falls.
I n the years to come, high rates of economic
growth will be needed in developing countries to
provide their rapidly growing populations with
improved living standards. If current trends in
energy and economic growth continue, commercial
energy consumption in the developing countries
could more than double over the next 40 years
according to most projections. Supplies of biomass
fuels would also need to increase substantially to
meet the needs of growing rural populations and the
urban poor.
This prospect raises a dilemma. On the one hand,
increases in energy supplies on this scale could
severely strain financial resources in the developing
countries. The energy sector absorbs a large share of
available foreign exchange and capital investment.
Consequently, energy supply policies have fa r-
-reaching impacts on other development priorities. In
many developing countries, financial resources may
not be adequate to increase commercial energy
supplies on the scale projected above.
On the other hand, inability to supply needed
energy can frustrate economic and social develop-
ment. Already in many countries, the unreliability
and poor quality of energy supplies lead to major
costs to the economy through wasted materials,
stoppage of operations, and investment in standby
equipment.
Energy prices are a key factor in the development
of a country's energy supply infrastructure, through
their impacts on the amount of energy used in the
economy, the technologies adopted, and, in some
cases, the direction of industrial development. En-
ergy prices in developing countries are typically
subject to price regulation throughout the distribu-
tion chain. The average level of energy prices,
particularly in the electricity sector, are reported to
be too low in many countries to ensure the sector's
financial viability.
Although commercial fuels attract the most policy
attention, two-thirds of the developing world's
population live in rural areas with low standards of
living based on low-resource farming. This popula-
tion has little access to commercial fuels and relies
largely on traditional sources of energy, gathered
and consumed locally, and animal and human
energy, often used at very low efficiencies. The main
form of traditional energy used is wood, an increas-
ingly scarce and unsustainable resource. This im-
poses a special hardship on those-mainly women
and children-responsible for gathering it. Dung
and crop wastes, the other forms of energy widely
used for cooking when wood is not available, have
alternative uses as soil nutrients.
Economic Development and Its
Impact on Energy
The pace of economic growth and level of
economic activity have major impacts on the energy
sector. From 1965 to 1987, for example, the econo-
mies of the developing countries grew at an annual
average of 5.3 percent, and their consumption of
commercial energy grew by just over 6 percent. 1
These energy growth rates were higher than those in
the industrial countries over the same period. As a
result, the developing countries' share of global
commercial energy consumption also rose-from 17
'World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 1989), pp. 167 and 173.
33-718
90
-25-
26 % Energy in Developing Countries
percent in 1973 to 23 percent in 1987. 2 High rates of
economic growth will continue to be needed in the
developing world to provide the rapidly growing
population with improved living standards.
Rising Populations
The past two centuries have witnessed sharply
accelerated growth in the global population (see
figure 2-1). The largest additions to global popula-
tion have been in the developing countries and have
occurred primarily in the past 50 years (see figure
2-2). Box 2-A discusses factors affecting population
growth.
Although current projections of global population
growth over the next 35 years differ (see figure 2-3),
there is consensus on several major points:
• The world's population is projected to increase
despite assumptions of continued declines in
fertility rates. The rate of increase in global
population, while lower than in the past, still
represents a large increase in numbers of
people. World Bank estimates, 3 which are
similar to both the United Nations medium
projection and the U.S. Department of Agricul-
ture projection, project an increase in global
population from 5.3 billion in 1990 to 8.4
billion in 2025, an increase of 3.1 billion.
• Virtually all of the increase will come from the
developing countries. According to the World
Bank projection, for example, population
growth in the industrial countries-i.e., nations
in the Organization for Economic Cooperation
and Development (OECD), the U. S. S. R., and
Eastern Europe-is expected to add only about
125 million, or about 4 percent of the global
increase (see figure 2-4). The population of the
developing countries is estimated to rise from
its present level of 41 billion to 7.1 billion in
2025, increasing their share of world popula-
tion from 77 to 88 percent (see figure 2-4).
• Population growth in China is projected to be
quite moderate, as current low rates of growth
are assumed to be maintained. Projections of
China's population growth are critical because
of its large share of the global total.
• According to World Bank projections, the
biggest increases in population are predicted to
12
10
8
e
4
Figure 2-1— World Population Growth, 1750-2100
in Industrial and Developing Regions
Population in billions
1900
1950
2000
2050
2100
SOURCE: Thomas Merrick, Population Reference Bureau, "World Popula-
tion in Transition," Population Bulletin, vol. 41, No. 2, April 1986,
update based on United Nations 1989 projections.
Figure 2-2 — Average Annual Increase in Population
Per Decade in Industrial and Developing Regions,
1750-2100
Population in millions
10 ~ -
Developing regions
Industrial regions
40-
^^^gHH
'201T -
1750
2100
SOURCE: Thomas Merrick, Population Reference Bureau, "World Popula-
tion in Transition," Population Bulletin, vol 41, No. 2, April 1986,
update based on United Nations 1989 projections.
come from Africa and Asia, which will account
for 30 and 58 percent respectively of the total
global increase. In Asia the large addition to
population derives from the existing large
population base; rates of population growth are
relatively low. In Africa, on the other hand, the
primary cause is the rapid increase in popula-
tion that, despite the relatively low population,
increases its share of the total population from
a current 12 percent to 19 percent in 2025.
ZWorld Energy Conference, Global Energy Perspective 2000-2020 14th Congress, Montreal 1989 (Paris: 1989), Table 2.
3 Rodolfo A. Bulatao, EduardBos, Patience Stephens, and My X Vu, Europe, Middle East, and Africa (EMN) Region Population Projection: 1989-90
Edition, Population and Human Resources Department, working paper 328 (Washington, DC: World Bank), November 1989, table 5.
Chapter 2~Energy and Economic Development »27
Figure 2-3-Historical and Projected Global
Population, 1950-2025
Population in billions
Figure 2-4-Projected Shares of Global Population,
2000 and 2025
U.N. (low, reed, high)
World Bank
/
2 -
1950
Variations among these projection arise
from differences in projected future fertility rates,
in turn are based on many assumptions, such as the
effects of government policies, GNP per capita,
literacy, and sociocultural factors.
1975
2000
2025
SOURCE: Office of Technology Assessment, 1990, based on data in
United Nations, World Population Prospects 1988 (New York,
NY: United Nations, 1989); Rodolfo Bulatao et al., Europe,
Middle East, and Africa (EMN) Region Population Projections,
1989-90 Edition, World Bank, Population and Human Re-
sources Department, Washington, DC, working paper series
328, November 1989.
Though there will no doubt be some unforeseen
divergence from these population paths (changes in
fertility rates are difficult to predict; the impact of
the AIDS epidemic in Africa on fertility rates and
population growth is unknown), it is clear that there
will be a large increase in the world's population in
the decades ahead, accompanied by a powerful
upward pressure on energy consumption. Even with
no increase in per-capita energy consumption, the
predicted rise in the global population by 2025
implies a 75 percent increase in total commercial
energy consumption.
Higher Living Standards
The major development challenge is to provide
higher standards of living for the rapidly rising
populations of the developing world. This task is all
the more urgent because of the declining levels of
per-capita income in many of the countries of Latin
America and Africa in recent years. Given the
projected rise in developing country populations—
an annual average of about 1.6 percent over the next
m Developing regions
I 1= industrial regions I
l i
World Bank projections
23%
1990
20%
2000
Year
16%
2025
SOURCE: Rodolfo Bulatao et al., Europe, Middle East, and Africa (EMN)
Region Population Projections, 1989-90 Edition, World Bank,
Population and Human Resources Department, Washington,
DC, working paper series 328, November 1989.
35 yearn-a rise in average per-capita incomes of,
say, 3 percent per year implies economic growth
rates of around 4.6 percent annually."
It may not be easy to achieve such rates of growth.
The current indebtedness of many developing na-
tions has added to the already difficult tasks of
economic management, and threatens to jeopardize
prospects of attaining even modest improvements in
standards of living. The foreign debt of developing
countries increased rapidly in the 1970s and 1980s
and in early 1989 was estimated at about $1.3
trillion. 5 As a result of this increase and the rise in
interest rates, debt service as a share of total exports
of goods and services is now double what it was in
the early 1970s. 6
Changes in Energy Consumption With
Economic Development
The economic expansion necessary to achieve
higher standards of living for the increasing popula-
tion of the developing world would be expected to
4 The World Energy Conference "moderate" growth rate projection is based on annual economic growth rates of 4.4 percent. Average economic
growthrates of 5.3 percent annually are assumed in the series of projections in Alan S. Marine and Leo Schrattenholzer, "InternationalE nergyWorkshop:
Overview of Poll Responses,' Stanford University International Energy Project, California, July 1989. The Intergovernmental Panel on Climate Change,
"Appendix Report of the Expert Group on Emissions Scenarios (Response Strategies Working Group Steering Committee, Task A)," April 1990,
assumes high economic growth rates for the different developing regions of 4 to 5 percent annually, and 2.2 to 3.0 percent annually for the low-growth
case.
'United Nations Development Programm^ u^man Development Report 1990 (New York, NY: Oxford University Press, 1990), p. 79.
'These principal and interest repayments are now much higher than new disbursements of long-term debt to developing countries. The net transfer
or outflow of resources from the developing countries amounted to $38 billion in 1987, compared with a net inflow of $35 billion in 1981. See World
Baok,World Development Report 1989 (Washington DC: Oxford University Press, 1989), p. 18.
28 • Energy in Developing Countries
Box 2-A — Factors Affecting Population Increase
Rates of population growth are determined by the balance between birth and death rates. Historically, death
rates were the first to decline, due to improvements in nutrition and sanitation, and medical advances such as
vaccines. If death rates decline, but birth rates remain the same, population increases. This is what happened in the
presently developed world from 1750 to about 1900. Around that time, however, birth rates started to fall, resulting
in an overall reduction in the rate of population increase. This process, the lagged adjustment of birth rates to the
prior decline in death rates, is known as the "demographic transition. " In the developing world, the demographic
transition is far from complete: death rates have fallen dramatically-though they are still higher than in the
industrial countries-but birth rates remain high, leading to a continued rapid rate of increase in total population. 1
Future trends in population will similarly depend on the balance between death and birth rates. For the
developing countries, opportunities still exist to reduce death rates through improvements in medicine and public
health, and further declines are likely and desirable. On the side of birth rates, there is much greater uncertainty over
future trends. Birth rates are falling in the developing countries, from 41 crude births per thousand population in
the mid-1960s to 30 per thousand at present. Birth rates in the developing countries, however, are still well above
death rates, and more than twice the birth rates in the industrial countries (currently 13 per thousand).
The number of births depends on three factors: fertility rates, the age structure of the population, and the size
of the population base.
The fertility rate is defined as "the number of children that would be born to a woman if she were to live to
the end of her childbearing years and bear children at each age in accordance with prevailing age-specific fertility
rates." Fertility rates in developing countries have fallen steadily, and in some cases sharply, in recent years. For
the developing countries as a whole, they fell from 6.1 in 1%5 to 4 in 1987, with particularly sharp declines in China
and India, Sri Lanka, Korea, and several Latin American countries. Despite this drop, they are still much higher than
in the industrial countries. There are, however, exceptions to this declining trend; fertility rates have not changed
in sub-Saharan African countries, and in some of these countries the rates appear to have risen.
Fertility rates are projected to continue declining until the end of the century, when they would be 3.3 compared
with the current 4. While this assumption seems reasonable in the light of historical trends, the determinants of
family size are not clearly known, and there is inevitably some degree of uncertainty over such assumptions. Broadly
speaking, fertility rates decline as levels of economic and social development and urbanization rise, women's
education improves, and knowledge about family planning spreads. The connections between these factors are not
well-established, however, as they are highly correlated, and it is therefore difficult to disentangle the impact of any
single determinant. A higher share of the population living in urban areas, other things equal, maybe of particular
importance in lowering fertility rates. In rural farming communities, many benefits accrue to a large family. Children
provide farm labor— from an early age children are able to perform simple farm chores. Children can also provide,
in the absence of social insurance, some guarantee of old age security for parents. These benefits of a large family
are not so evident in the urban context, where they may also be outweighed by the financial costs of supporting a
large family.
Other factors also influence fertility rates. Cultural and religious factors can lead to higher family size than
would otherwise be predicted by indicators of social and economic development and urbanization. Algeria, Libya,
Iran, and Iraq, for example, have fertility rates near or over 6 despite their relatively high per -capita incomes. On
the other hand, aggressive government policies to restrict families can lead to lower family sizes than predicted by
other social and economic indicators. For example, fertility rates in China, a low-income country, fell dramatically
from 6 in the mid-1960s to 2.4 in 1987— lower than the rate for the industrial countries 20 years ago-due largely
to strong government policy.
With a given fertility rate, the number of births will be higher if a larger share of the population is in the
reproductive age group. In the developing countries, young people comprise a higher share of the population. This
population structure gives a much greater "population momentum" (the tendency for population growth to
continue even after fertility rates have fallen to the replacement level). In the developing countries, the rising share
of women of childbearing age in the population will continue to exert strong upward pressure on the population,
despite the expected drop in fertility rates.
Chapter 2~Energy and Economic Development *29
The size of the population base is the third determinant of population growth. The highest fertility rates exist
in the developing countries, which already have by far the largest share (77 percent) of the global population. This
means that larger numbers are being added to the world population than would be the case if the high fertility rates
applied to the industrial countries, which comprise only 15 percent of the global population.
1 World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 1989), pp. 164- 165.
lead to comparable increases in energy consump-
tion. In addition, several factors inherent in the
development process tend to cause commercial
energy use in developing countries to rise more
rapidly than the gross national product.
First, most of the people in developing countries
now rely primarily on traditional biomass fuels-
wood, crop wastes, animal dung-for their energy
needs. These fuels are often difficult and time-
consuming to gather, are inefficient and awkward to
use, and can cause significant environmental dam-
age. Similarly, most people in developing countries
rely primarily on human and animal muscle power
for doing their work, despite its low efficiency and
limited output. People who are dependent on these
traditional forms of energy will turn to commercial
fuels and technologies if and when they are available
and affordable. Thus, the share of traditional fuels in
total energy consumption falls sharply as develop-
ment proceeds. In the low-income African countries,
traditional fuels account for as much as 90 percent of
total energy use. In the middle-income developing
countries their share falls typically to under 20
percent.
Second, most developing countries are now build-
ing their commercial, industrial, and transportation
infrastructures. This requires large quantities of
energy-intensive materials such as steel and cement.
As a result, energy use in the near- to mid-term
increases faster than income.
Third, developing countries are experiencing
rapid urbanization. Urbanization has profound ef-
fects on the amount and type of energy consumed. 7
As industry and the labor force become more con-
centrated in urban areas, transportation needs grow.
Food and raw materials are hauled longer distances,
and finished products are marketed over a wider
area. Urban households purchase a larger share of
their total needs from outside the family, compared
with rural households, and commercial providers of
goods and services are more likely to use modern
fuels. Scarcity of space in cities encourages the
substitution of modern, compact energy forms for
the bulkier biomass fuels. Finally, the growing food
needs of the cities encourage changes in agricultural
technology, which usually involve increased use of
modern fuels and energy-intensive fertilizers.
Fourth, modern manufacturing technologies and
materials have significantly lowered the real cost of
many consumer goods-from radios to refrigerators-
compared with costs a generation ago, and global
distribution systems have increased their accessibil-
ity. People in developing countries can thus pur-
chase many consumer goods at a far earlier point in
the development cycle than did people in today's
industrial countries. This could increase energy use
in the near- to mid-term both to produce the
materials for consumer goods and— particularly for
those that are intensive energy users, such as
motorcycles, cars, air conditioners, and refrigerators—
to operate them.
On the other hand, there are factors that may
counterbalance these trends and significantly mod-
erate the rapid increase in energy demand.
First, the high cost of developing national energy
infrastructures and of purchasing energy to support
growing energy demands could potentially sharply
limit economic growth. This possibility is highly
undesirable given current low, and in many cases
declining, living standards in developing countries.
Second, the expected growth i n energy use i n
developing countries could be reduced through
efficiency improvements. Energy is now used much
less efficiently in developing countries than in
industrial countries. Traditional fuels and technolo-
gies are often much less efficient than modern ones:
for example, the efficiency of a typical wood-fueled
cooking stove is just one-fourth that of a modern gas
range. Moreover, the efficiency of energy use in the
Donald W. Jones, "Urbanization and Energy Use in Energy Development" Energy Journal, vol. 10, No. 4, October 1989.
30 • Energy in Developing Countries
modern sector in developing countries is often far
lower than that commonly achieved in the industrial
countries. If developing countries adopt the most
efficient technologies now available, they might
achieve average energy efficiencies that are higher
than those in industrial countries that have a large
installed base of older and less efficient infrastruc-
ture and equipment.
Third, continued economic development is at
some point accompanied by structural changes that
shift investment from energy-intensive infrastruc-
ture (roads, buildings, etc.) to consumer goods
(refrigerators, cars, etc.) and finally to less material-
intensive but higher value-added goods such as
personal services and electronics.
Energy use in developing countries will depend
on the net impact of these opposing factors. At low
levels of development the first set of factors predom-
inates, and commercial energy consumption typi-
cally rises much faster than gross domestic product
(GDP). Figure 2-5 compares per-capita commercial
energy consumption with per-capita GDP for se-
lected countries, ranging from lowest to highest
income. 8 As this figure suggests, within the poorest
countries commercial energy consumption rises
faster than per-capita GNP; in the middle-income
countries they rise at about the same rate; and at the
highest levels of income, the increase in total
commercial energy consumption is less than the
increase in per-capita GNP.
It is often argued'that GNP per capita, a measure
of the val ue of economi c output i n rel ati on to
population size, is an inadequate and misleading
indicator of standards of living and well-being. 10 An
index recently developed by the United Nations
Development Programme, the Human Development
Index (HDI)," incorporates both economic and
social factors. This index is based on three indica-
tors: life expectancy at birth, adult literacy, and
per-capita purchasing power. 12 The first two are
sensitive to social conditions in a country and in
addition reflect underlying conditions of income
distribution. Average purchasing power in a country
gives some indication of material standards of
living. At low values, the HDI also shows a close
positive association with commercial energy con-
sumption (see figure 2-6). Higher levels of HDI,
however, can be achieved with a wide range of
commercial energy consumption.
Commercial energy is only part (and for the
poorest countries, a very small part) of total energy
consumption. If estimates of traditional fuel con-
sumption are included with commercial fuel to
represent total energy consumption, the association
between per-capita energy consumption and GNP
remains close, but at lower income levels the slope
is less steep than in figure 2-5. 13 Adding commercial
and traditional energy together to make total energy
consumption does not, however, take into account
the lower conversion efficiencies of traditional
energy compared with commercial energy. If tradi-
tional energy consumption were expressed in com-
8 The relationship between GDP per capita and commercial energy consumption shown in figure 2-5 is consistent with other studies. A per-capita
income elasticity for fossil fuels of about 1.5 is given for 13 industrial and developing countries in Gerald Leach et a\.JZnergy and Growth (London:
Butterworths, 1986),p. 25. That is.agiven increase in per-capita GNP between countries(purchasing power parity) is associated with a 50 percent higher
increase in consumption of fossil fuels. Another study, based on 100 countries, reports a per-capita income elasticity of 1.26 (i.e., a given increase in
per-capita GNP between countries is associated with a 26 percent higher increase in commercial energy consumption). See B.W.Ang, "A
Cross-Sectional Analysis of Energy Output Correlation" Energy Economics (London: Butterworths, 1987), table 3, p. 280. This elasticity is based on
market exchange rates; comparison of purchasing power parity GNP data increases the elasticity to 1.8. These elasticities are based on cross-sectional
studies (i.e., intercountry comparison at a given point in time, rather than developments in an individual country over time), which are considered to
give a more accurate picture of the long-term relationship between energy consumption and economic growth.
*For example, see Carlos Andrea Perez, Towards a New Way to Measure Development (Caracas, Venezuela: Office of the South Commission in
Venezuela, 1989).
10 GNP may be an inadequate measure of social well-being, but it nonetheless correlates strongly with many social indicators. See Partha Dasgupta,
"Wellbeing and the Extent of ItsRealization in Poor Countries," The Economic Journal, supplement (Cambridge, MA: Royal Economic Society Basil
Blackwell, 1990), pp. 1-32. The argument is made here that GNP per capita also correlates closely with political and civil rights.
"United Nations Development Programme, Human Development Report 1990, op. C it., footnote 6, pp. 11-16.
12 The values of achievement for these indicators were Japan' s life expectancy at birth of 78, an adult literacy rate of 100 percent, and the average official
"poverty Une" income in nine industrial countries, adjusted to take into account purchasing power, of $4,861 per capita
13 Leach, op. c ' l -> footnote 9, p, 25, gives per-capita income elasticities for total energy at about 1.2, lower than the 1 .5 for fossil fuels alone. B.W.
Ang, "A Cross Sectional Analysis of Energy Output Correlation" op. cit., footnote 9, reports income elasticities of traditional energy consumption of
-0.95 (i.e., a 10 percent increase in GNP per capita is associated with a 9.5 percent decline in traditional energy consumption) and for total energy
consumption (commercial and traditional) of 1.05.
Chapter 2™Energy and Economic Development .31
1000
100
Figure 2-5-Commercial Energy Consumption and Economic Development
in Selected Countries
Gigajoules per capita
10
0.1
.22
■ 23
-i 19
17.. 18
5 6 11
.,'% : 1
■ 2
■ 3
.13
i ; i ■
.20
.16
1 Ethiopia
2 Nigeria
3 Kenya
4 India
13 Algeria
14 Peru
15 Costa Rica
16 Brazil
5 Zimbabwe
17 Korea
6 Egypt
7 Cameroon
18 Argentina
19 Venezuela
8 Indonesia
20 Mexico
9 Morocco
21 Japan
10 Philippines
22 France
11 China
23 W. Germany
12 Thailand
24 United States
__
J_
_L
I
I
o 2 4 6 8 10 12
Gross domestic product per capita (thousand dollars) (ppp)
14
SOURCES: United Nations, 1986 Energy Statistics Yearbook(New York, NY: United Nations, 1988), table 4; Robert
Summers et al., "A New Set of International Comparisons of Real Product and Price Levels, Estimates for
130 Countries, 1950-1985," Review of Income and Wealth, Series 34, No. 1, March 1988.
mercial fuel equivalent, "the increase in energy
consumption in relation to GNP might be somewhat
greater than the increase in total energy (an energy
elasticity of just over 1.0), but less than the increase
in commercial energy alone (1.5 or more).
The inclusion of traditional energy, though neces-
sary to provide a more complete picture of the
relationship between economic growth, social de-
velopment, and energy use, raises problems of its
own. First, the measurement of traditional fuels is
difficult and prone to underestimation. For example,
for Indonesia, Malaysia, the Philippines, and Thai-
land, estimates by the United Nations Food and
Agricultural Organization, the Asian Development
Bank, and the World Energy Conference-three
frequently quoted sources of data on traditional
fuels-are found in almost all cases to be consider-
ably lower than in other cou.ntry-specific studies. 15
Second, when considering traditional sources of
energy, it is difficult to know whereto draw the line.
Animate forms of energy are important in most
developing countries, particularly the poorest. If the
biomass fed to bullocks to provide plowing and
irrigation services were included, the amount of
traditional energy consumed would increase sub-
stantially. Further, if the large amounts of biomass
burned in preparing soil for cultivation in slash and
burn agriculture were included, per-capita energy
use where shifting agriculture is practiced could
conceivably be as high or higher than in the
industrial countries. 16
Although GNP growth is an important determi-
nant of energy consumption, it is not the only one.
Table 2-1 shows commercial and total energy
consumption relative to GNP for different levels of
income. If the amount of energy consumed were tied
to the level of output or development, the energy
intensities of all countries would be the same. As
these figures show, however, there are considerable
variations in energy intensity (the ratio of energy
consumption to GNP) at all levels of development.
Among the industrial countries, energy intensities
vary widely. And among the developing countries,
energy intensities of countries with the same level of
14 B.W.Ang, 'A Method for Estimating Non-commercial Energy Consumption in the Household Sector of Developing Countries," Energy (London:
Pergamon Press, 1986), p. 423, table 8.
15&, op.cit., footnote 15, p. 423, table 8 -
16 See Kirk R - Smith, "The Biofuel Transition," Pacific and Asian Journal of Energy, vol. 1, No. 1, January 1987, p. 18, figure 5.
32 • Energy in Developing Countries
Figure 2-6-Commercial Energy Consumption and Social Development
in Selected Countries
Giq
ajoules per
capita
66
o 1
. 24
300
1 Etmopia
13 Algeria
2 Nigeria
14 Peru
3 Kenya
15 Costa Rica
4 India
16 Brazil
250 1
5 Zimbabwe
17 Korea
6 Egypt
18 Argentina
7 Cameroon
19 Venezuela
200
8 Indonesia
20 Mexico
9 Morocco
21 Japan
• 2 3
10 Philippines
22 France
11 China
23 W. Germany
m22
lbO
12 Thailand
24 United States
• 21
• 19
100
18
50
■ 13
20 .,7
, 16
7 ■ r3 ■ 5
11 •
.14 a15
I >
2 4 - ~ m ~ 9
8 10m
■ 12
o 0.2 0.4 0.6 0.8 l 1. 2
Human development index
SOURCES: United Nations, 1986 Energy Statistics Yearbook(New York, NY: United Nations, 1988), table 4; United
Nations Development Programme, Human Development Report 1990 (New York, NY: Oxford University
Press, 1990), p. 79.
Table 2-I — Energy Intensities in Selected Countries
Gross domestic Total energy Commercial energy
product, 1985 consumption per GDP consumption per GDP
Country $PPP' GJ/$1,000 Index, U.S.= 100 GJ/$1,000 Index, U.S.=1QO
Ethiopia 304 29.6 123 2.3 10
Nigeria 565 24.8 103 9.0 38
Kenya 603 29.9 124 5.0 21
India 775 15.5 64 11.4 48
Zimbabwe 954 30.4 126 22.3 94
Egypt 1,080 21.3 88 20.6 87
Cameroon 1,180 16.9 70 9.1 38
Indonesia 1,269 12.6 52 6.6 28
Morocco 1,284 7.8 32 7.3 31
Philippines 1,352 11.1 46 6.7 28
China 1,489 15.4 64 14.3 60
Thailand 1,896 12.7 53 7.0 29
Algeria 2,133 17.8 74 17.4 73
Peru 2,333 9.9 41 8.1 34
Costa Rica 2,712 16.2 67 8.3 35
Brazil 3,164 16.4 68 10.7 45
Korea 3,381 16.6 69 16.1 68
Argentina 3,640 17.0 71 15.9 67
Venezuela 3,723 29.8 124 29.9 126
Mexico 3,987 13.3 55 12.5 53
Japan 9,739 12.7 53 12.7 54
France 10,032 16.1 67 16.0 67
Germany 10,959 17.0 70 16.9 71
United States 12,787 24.1 100 23.7 100
a ppp refers to purchasing power Parity.
SOURCES: United Nations, 1986 Energy Statistics Yearbook (New York, NY: United Nations, 1988), table 4 for energy consumption data. Robert Summers
et al., "A New Set of International Comparisons of Real Product and Price Levels, Estimates for 130 Countries, 1950-1985," Review of Income
and Wealth, Series 34, No. 1, March 1988 for gross domestic product, purchasing power parity.
Chapter 2™Energy and Economic Development %33
development (measuredly GNP per capita) can vary
more than fourfold. 17
Variations in energy intensity are also evident in
individual countries or groups of countries over
time, as shown in figure 2-7. In the years immedi-
ately before 1973, global energy intensities rose.
After 1973, there was a sharp drop in the industrial
countries, in contrast to a continued rise in the
developing nations, if at a lower pace. Experience
within the developing countries varied considerably,
again testifying to the importance of factors other
than economic growth in determining energy con-
sumption.
Differences in rates of urbanization and industrial
structure account for part of the variation in energy
intensities. Countries with a large share of energy-
intensive industries, such as steel, paper, chemicals,
and aluminum, will tend to have higher energy
intensities than countries with few energy-intensive
industries. Nigeria, Egypt, Algeria, and Venezuela
(with large oil drilling and refinery operations) and
Argentina, Korea, Zimbabwe, and China (with large
metals industries, mainly iron and steel) all have
relatively high energy intensities.
The impact of changes in industrial mix is also
evident over time. About half of the post-1979
decline in energy intensity in China (which fell by 40
percent between the late 1970s and the late 1980s)
can be ascribed to the limits on the expansion of
heavy industries and to the promotion of light, and
often export-oriented, manufactures (e.g., textiles,
consumer electronics, processed foodstuffs, and
plastics) .18 The equally dramatic decline in Korea's
energy intensity is also due in large part to the
changing industrial mix. Although the output of
heavy industries in Korea rose sharply in these years,
production of less energy-intensive industries such
as machinery and transport equipment grew even
more rapidly, resulting in a declining share of heavy
industry in the total.
Energy intensities are also influenced by the
technologies used throughout the economy. For
example, the older generation of coal-burning tech-
nologies still used in developing countries are much
Figure 2-7 — Energy Intensity and Economic
Development, 1960-2000
Barrels of oil equivalent per $1,000 of goods and services
'- Industrial nations
'Developing nations
World average
SOURCE: World Bank, Industry and Energy Department, "Energy Issues
in the Developing World," Energy Series Paper No. 1, February
19s8.
less efficient than modern technologies. Conse-
quently, the coal-burning countries (India and China)
are more energy-intensive than countries largely
reliant on oil and gas.
Some of the factors that determine energy con-
sumption (e.g., the size of the country and the
location of natural resources in relation to industry
and major markets) are country-specific, but others
can be affected by policy decisions.
Projected Energy Consumption in
Developing Countries
In the past, the three factors discussed in the
preceding section— rapidly rising population, high
economic growth rates to provide improved stand-
ards of living, and structural change as development
gets underway-have been associated with rapid
rates of increase in commercial energy consumption
in developing countries. If these trends continue,
increases in commercial energy consumption in
developing countries could be very large. Table 2-2
illustrates some of the current projections. The
synthesis of a wide range of projections shown in
this table suggests that commercial energy con-
sumption in the developing world (including here
OPEC) in 2010 could be 2.5 times higher than it was
in the base year, 1985, an annual rate of increase of
17 From the limited sample shown in table 2.1, there does not appear to be a systematic tendency for energy/GNP ratios to rise as levels of development
rise; the energy/GNP ratios of the advanced developing countries are very similar to those of Europe and Japan. However, Ang, op. cit., footnote 9, using
a wider sample, shows a commercial energy/GNP ratio with respect to GNP per capita at 0.80. That is, a 10 percent increase in GNP is associated with
an 8 percent increase in the energy/GNP ratio.
18 VaclavSmil,' 'China's Energy: A Case Study, « contractor report prepared for the Office of TechnologyAssessment, April 1990.
34 % Energy in Developing Countries
3.8 percent. China accounts for more than one-third
of the projected increase.
The World Energy Conference forecasts a some-
what slower rate of growth in commercial energy
consumption in the developing world, an annual
average increase of 3.3 percent. By 2020, however,
consumption of commercial energy in the develop-
ing world would be three times higher than in 1985,
and consumption of traditional fuels about 25
percent higher (see figure 2-8). Population growth
and rising standards of living each account for about
half of the total increase. 19
The Energy Sector and the
Macroeconomy
Energy is widely recognized as a key economic
sector in developing countries. Reliable and afford-
able supplies of energy make major contributions to
economic and social development; conversely, in-
adequate or unreliable energy supplies frustrate the
development process.
Financing Energy Supplies
Most developing countries (59 out of 80) are net
energy importers, relying on imports for virtually all
domestic commercial energy consumption; for many
countries, oil imports represent 30 percent or more
of total export earni ngs. 20 The share of energy
imports in the total export earnings of developing
countries has fluctuated widely in recent years as oil
prices have changed. Such wide fluctuations are
highly disruptive to energy planning and economic
development.
Many developing countries emphasize domestic
production of energy. Investments in highly capital-
intensive energy supply systems often represent a
major share of the total investment budget (see table
2-3), accounting in some countries for over 40
Figure 2-8-Historical and Projected Energy
Consumption in Developing Countries: World
Energy Conference "Moderate" Projections
Exajoules
2S0
150
100
50
1970 1980 1990 2000 2010 2020
SOURCE: World Energy Conference, Conservation and Studies Commit-
tee, Global Energy Perspective 2000-2020, 14th congress,
Montreal 1989 (Paris: 1989).
percent of all public in vestment. 21 As the predomi-
nant claimant on scarce capital resources, develop-
ments in the energy sector therefore have a major
impact on the amounts of investment available for
other economic and social development.
The scale of future investment demands for the
energy sector in developing countries is projected to
be very large. The World Bank, for example,
estimates that investments of $125 billion annually
(twice the current level) would be needed in
developing countries to provide adeguate supplies of
electricity. According to a World Bank estimate
annual average expenditures on commercial energy
supply facilities for developing countries, electricity
accounts probably for one-half of the total; oil,
including refineries, accounts for about 40 percent;
and natural gas and coal 5 percent each. These
expenditures do not include investment in small-
scale renewable or energy conservation.
15 The Intergovernmental Panel on Climate Change has undertaken projections of energy consumption in developingcountriesin"AppendixReport
of the Expert Group on Emissions Scenarios (Response Strategies Working Group Steering Committee, Task A)," April 1990. Several scenarios are
provided, with different rates of economic grow @ emission coefficients, and policies. The high economic growth (growth rates similar to the other
studies quoted here), low emissions scenario forecasts a threefold increase in developing world energconsumption between 1985 and 2025, which is
reasonably similar to the World Energy Conference forecast. The high growth, high emissions scenario is similar to the combined forecast results in Alan
S. Marine and Leo Schrattenholzer, International Energy Workshop: Overview of Poll Responses (Palo Alto, CA: Stanford University, International
Energy Project, July 1989).
Energy imports and debt service together account for over one-third of total export earnings in middle-income countries, and almost 40 percent in
lower-income countries (excluding China and India).
21 These & @ cover investments in the energy sector by public entities. They may include some small amounts of investmenin energy conservation,
but virtually all is in energy supplies. They do not include private investment in backup equipment.
22 World Bank, "Capital Expenditure for Electric Power in the Developing Countries in the 1990' s," World Bank Industry and Energy Department
Working Paper, Energy Series Paper No. 21, Washington DC,February 1990.
Chapter 2™Energy and Economic Development ,35
Table 2-2--Commerciai Energy Consumption, 1985 and Projections for 2010 (EJ=Exajoules)
Increase in
Share of Share of AARG," consumption, Share of
1985 total 2010 total 1985-2010 1985-2010 increase
(EJ) percent (EJ) percent percent (EJ) percent
Developing countries 69.26 23.3 175.156 34.5 3.8 106.30 50.2
China 21.91 7.4 59.54 11.7 4.1 37.63 17.8
OPEC 11.65 3.9 28.91 5.7 2.5 17.26 8.1
Non-OPEC developing
countries 35.70 12.0 87.11 17.1 3.6 51.41 24.3
OECD 155.83 52.4 215.03 42.2 1.3 59.20 27.9
United States 73.87 24.8 98.47 19.3 1.2 24.60 11.6
U.S.S.R. and
Eastern Europe 72.32 24.3 118.66 23.3 2.0 46.34 21.9
Total 297.41 100.0 509.25 100.0 23 211.85 100.0
a AARG- annual average rate of growth.
SOURCE: Alan S. Marine and Leo Schrattenholzer, International Energy Workshop: Overview of Poll Responses (Palo Alto, CA: Stanford University,
International Energy Project, July 1989).
Table 2-3-Estimated Annual Energy investment as a Percentage of
Annual Total Public investment During the Early 1980s
Over 40 percent
30-40 percent
20-30 percent
10-20 percent
0-10 percent
Argentina
Ecuador
Botswana
Benin
Ethiopia
Brazil
India
China
Egypt
Colombia
Pakistan
Costa Rica
Ghana
Korea
Philippines
Liberia
Jamaica
Mexico
Turkey
Nepal
Morocco
Nigeria
Sudan
SOURCE:
About one-half of total estimated energy supply
investments are projected to be in foreign ex-
change. 23 The foreign exchange component for oil
and gas is typically high (about two-thirds of the
total), as much of the equipment must be imported.
On the other hand, foreign exchange costs for coal
development are low (about one-quarter of the total),
mainly because the major coal-using countries, India
and China, manufacture coal industry equipment
domestically. The projected share of foreign ex-
change in electric power varies widely according to
country. In countries with developed industrial
sectors, the share may be between 5 and 10 percent,
but in countries that import all their generating
equipment, the share rises as high as 70 Percent. 24
Achieving these high levels of foreign resources
for the energy sector investment poses immense
challenges. Though most countries are likely to
experience difficulties, the issues will vary from
country to country. The poorest countries are highly
dependent on confessional aid (which accounted for
80 percent of their total external borrowing for the
energy sector in 1975-80). Their success in acquir-
ing funds will depend on the extent of the increase
in confessional flows. On the other hand, the
middle-income countries depend mainly (80 per-
cent) on export-related and private financial flows
for their external financing of energy investments.
The situation is particularly acute in highly indebted
developing countries.
The other half of the projected increase in
investment in the energy sector comes from domes-
tic resources, with a particularly high share in
electricity and coal. In many countries, however, the
financial situation of the power sector has deterio-
rated, as increases in costs have not been matched by
^External funding for the energy sector was projected to come largely (almost three-quarters) from private sources (supplier credit and private
commercial loans). Loans from multilateral agencies and bilateral aid accounted for about one-quarter.
^World Bank, The Energy Transition in Developing Countries (Washington, DC: 1983).
36 • Energy in Developing Countries
increased revenues. 25 The financial viability of oil
refinery operations in many countries is also com-
promised by the structure of petroleum product
prices. 26 The issue of domestic resource mobiliza-
tion is of particular importance for coal, where much
of the financial resources needed are local rather
than foreign.
There are indications that the developing coun-
tries are paying increased attention to resource
mobilization for the energy sector. Several countries
(e.g., Peru, Ecuador, and Colombia) are currently
opening more of their territories to oil exploration by
foreign firms. Requirements for government partici-
pation in oil development ventures are being re-
laxed. Improved fiscal arrangements providing for
the special characteristics of gas have been adopted
in Egypt, Tunisia, Pakistan, and elsewhere with a
dramatic increase in exploration specifically di-
rected at gas. 27
Energy Supply Reliability
J ust as the presence of reliable supplies of
high-quality energy can be a strong incentive to
economic development, so unreliable supplies can
discourage development and add substantially to the
cost of usable power. Electricity supplies in many
developing countries are characterized by frequent
service curtailments to customers, including black-
outs, brownouts, and sharp power surges. This can
have two types of impacts:
• Industries and offices are unable to operate,
production is lowered, and raw materials are
wasted. In China, for example, it is claimed that
electricity shortages and disruptions during the
1980's were responsible for idling at least 20
percent of the country's industrial capacity. 28
For the five public-sector steel plants in India,
it has been estimated that at 1986-87 operating
levels, irregular and restricted electricity sup-
ply resulted in increased electricity consump-
tion of over 216 gigawatthours at a cost of $10
million, and the poor quality of the electricity
resulted in additional consumption of 412
gigawatthours at a cost of $18 mi 1 1 ion. 29 More
generally, lost industrial output caused by
shortages of electricity in India and Pakistanis
estimated to have reduced GDP by about 1.5 to
2 percent. 30 Residential consumers are also
affected.
• Many consumers, both residential and indus-
trial, are obliged to invest in a variety of
equipment-voltage boosters, standby genera-
tors, storage batteries, kerosene lamps-in
order to minimize the impact of disrupted
supplies. Though no data are available, expen-
ditures on these devices are certainly substan-
tial, adding to the cost of providing usable
supplies.
Such supply constraints are usually associated
with electricity, but there are also shortages of other
sources of energy. Supplies of household fuels in
many countries (e.g., India) are notoriously intermit-
tent. This accounts for the existence of a wide range
of cooking systems in many households in order to
ensure against the shortage of any one fuel. Trans-
portation services are also subject to disruption
because of unreliable fuel supplies.
Energy Pricing and Demand
Management
Energy prices play a key role in energy sector
development, through their impacts on the amount
of energy used in an economy, the technologies
adopted, and in some cases, the direction of indus-
trial development. The effects on the energy infra-
structure are long term in nature, and often difficult
to reverse.
Energy pricing policy may have several objec-
tives: efficient allocation of resources, provision of
affordable supplies to consumers, reasonable returns
to energy producers, substitution between fuels for
^World Bank, "Review of World Bank Lending for Electric power, " World Bank Industry and Energy Department Working Paper Energy Series,
Paper No. 2, March 1988.
^Donald Hertzmark, "Energy Efficiency and Energy pricing in Developing Countries,' ' contractor report prepared for the Office of Technology
Assessment, May 1990.
^Theodore J. Gorton, ' 'Petroleum in the Developing World," contractor report prepared for the Office of Technology Assessment, July 1990.
^Vaclav Snail, "China's Energy: A Case Study, " op. cit., footnote 18.
Energy and EnvironmentalAnalysis, "Conserving Process Heat in Primary Industries of India anoChina/'contractorreportpreparedfor the Office
of Technology Assessment, April 1990.
^Arun P. Sanghvi, "Impacts of Power Supply Inadequacy in Developing Countries,"./oH™«f of Energy Policy (forthcoming).
Chapter 2~Energy and Economic Development »37
national security or environmental reasons, promo-
tion of regional development and industrial compet-
itiveness. 31 The weights of these different objectives
in the formulation of energy policy vary among
countries. The importance of the regional develop-
ment objective, for example, varies from country to
country depending on geographical configuration,
politics, and history. As in other aspects of economic
and social policy, however, there are several charac-
teristics of energy pricing that are shared by many of
the developing countries:
• Governments play a strong role in the commer-
cial energy sector. In virtually all developing
countries the electricity sector is government
owned, and in many countries the government
also owns the coal, and oil and gas sectors.
Regardless of the form of ownership, the
government typically regulates prices of energy
products, frequently at several levels of the
production and distribution chain.
• Social objectives are an important factor in
formulating energy pricing policies. As a basic
necessity of life, energy accounts for a substan-
tial part of total household expenditures. Gov-
ernments frequently aim to keep the typical
cost of household fuels-e. g., kerosene for
lighting and cooking, and in some cases elec-
tricity-low. The large number of poor in the
population also makes price stability an impor-
tant policy objective. Though social equity
issues are a major preoccupation in the pricing
of commercial fuels, prices of the most com-
mon form of energy used by the poor— wood
and charcoal-are usually not regulated.
• Economic objectives, notably, the desire to
encourage key strategic development sectors
including transportation and agriculture, are
also reflected in policies designed to promote
rural electrification or to keep diesel prices low.
Policies that keep key energy product prices low
can also produce adverse results. Revenues from
energy sales may be inadequate to cover the costs of
supplying the energy. This problem is especially
acute in the electricity generating sector in develop-
ing countries. One study 32 showed that in 30 out of
37 developing countries for which data were avail-
able, electricity tariffs were too low to generate the
revenues needed to cover total operating costs plus
allowances for equipment replacement or expansion
of the system. A survey of electric power projects
financed by the World Bank over a 20-year period 33
indicates a consistent decline in key financial
indicators as revenues from sales of electricity
lagged behind rising costs.
Petroleum prices are rather different. At present,
subsidies (defined here as prices significantly lower
than those charged in international markets) are
largely limited to oil -exporting (or at least oil-
producing) countries. Countries that import all their
supplies of petroleum products, and are therefore
obliged to pay current international prices for their
supplies, are generally unwilling to subsidize prices
on the domestic market. In the oil-exporting coun-
tries, however, despite sharp price increases in
recent years, several petroleum products— notably,
kerosene and heavy fuel oil-continue to be sold at
half the international price or less. Low domestic gas
prices, in combination with other factors, discourage
the development of gas resources and contribute to
the spectacularly high share of flared gas in develop-
ing count ries-47 percent of total production, com-
pared with 4 percent in the OECD countries.*
In both India and China, which together account
for 70 percent of all coal consumption in the
developing world, coal prices are kept below pro-
duction costs. In China, two-thirds of all coal
enterprises lost money in 1984. 35 ln the early 1980s
the World Bank estimated Coal India's losses at
$300 million on sales of $700 million. 36
31 For further discussion of the scope and objectives f energy pricing, see Mohan Munasinghe, Energy Analysis and Policy (London: Butteiwoiths,
1990); Lawrence J. Hill, Energy Price Reform in Developing Countries: Issues and Options (Oak Ridge, TN: Oak Ridge National Laboratory, August
1987), and Cotazon Siddayao, Criteria for Energy Pricing Policy (London: Graham and Trotman, 198S).
32 LawrenceJ. Hill, op. cit., footnote 31, pp. 2-10 and 2-20, table 2-3.
33 World Bank, "Review of World Bank Lending for Electric Power, " Industry and Energy Department Working Paper, Energy Series paper No. 2,
March 1988.
^MarkKosmo, Money toBurn? Tne HighCost of Energy Subsidies (Washington, DC: world Resources Institute, 1987), p. 14. Based on International
Energy Agency data.
35 Lawrence J. Hill, op. cit., footnote 31.
^Mark Kosmo, op. cit., footnote 34, p. 16.
38 • Energy in Developing Countries
I n some cases, the costs of energy supplies are
also higher than necessary. Factors such as excessive
staffing and poor management in the electricity
sector increase costs, and there are similar ineffi-
ciencies in the oil supply sector. Insofar as the
population is aware of these problems, they maybe
reluctant to agree to price increases that would in
effect subsidize the inefficiencies of the supply
system. Improved efficiencies on the supply side
might make increases in prices and tariffs more
palatable and also help to minimize the total cost to
consumers.
In addition to the general level of energy prices,
the structure of energy prices is of concern in both
the electricity and petroleum product markets. Major
differences in the prices charged for similar services-
as in the case of electricity-or for petroleum
products that can be substituted for each other-
have given rise to distortions in product demand.
The subsidization of some fuels (kerosene and diesel
fuels) for general economic and social reasons,
combined with high taxes on others (gasoline), leads
to shortages of the subsidized fuels, surpluses of the
highly taxed fuels, and capital investment decisions
made on the basis of energy costs that do not reflect
the cost of providing that energy.
In Thailand in the early 1980s, for example, price
differences between gasoline, diesel, kerosene, and
liquid petroleum gas (LPG) led to shortages and
black markets in kerosene and LPG; the diversion of
half of the total kerosene supply to the transport
sector to adulterate diesel fuel; dieselization of many
older vehicles by retrofitting a spark ignition engine
to use diesel fuel; widespread theft of diesel fuel; and
surpluses of gasoline as all vehicles used commer-
cially changed over to diesel. 37 Similar develop-
ments in other countries have contributed to serious
refinery imbalances. In recent years, Thailand has
moved to reform its petroleum product pricing
system, but wide price differentials persist in other
countries, including Indonesia and India.
Energy pricing decisions are often motivated by
the need to keep energy affordable for large popula-
tions of poor households. However, the practical
implementation of such policy is difficult. It is often
difficult to "target" disadvantaged groups. Energy
consumption surveys indicate that the use of com-
mercial fuels is concentrated among middle and
upper income households, rather than the poor who
rely mainly on wood and charcoal. Moreover, if
subsidy programs expand in scale, they can lead to
outcomes that penalize the very people they are
designed to help. Thus, the deteriorating revenue
situation of electricity systems, attributable in some
measure to subsidized tariffs, leads to declining
quality and availability of power supplies, which can
cause factories and workshops to stop operations,
thus increasing unemployment.
Though important, pricing is just one mechanism
for influencing energy demand. Others include
measures to inform consumers of cost-effective
opportunities to save energy, the imposition of
technical efficiency standards, and sponsorship of
energy-efficient technologies.
Developing countries, frequently aided by donor
countries and organizations, have made some prog-
ress in demand management and conservation. For
example, the Association of South-East Asian Na-
tions (ASEAN) countries (see box 2-B) have been
particularly active in conservation in both industries
and buildings. In addition, China has established
energy conservation technical centers, which have
contributed to the sharp decline in China's energy
intensity. In Brazil, energy-saving protocols have
been established with major industries. Korea has
conducted major audits of large companies. Traffic
management schemes, designed mainly to alleviate
congestion, but with an energy-saving bonus, have
been introduced in Brazil, Singapore, Thailand, and
Venezuela. New, more energy-efficient automobile
technologies have been introduced in India. And
several improved wood-burning stoves have been
introduced, at least one of which appears to have
enjoyed considerable success. 38 On the institutional
side, movements toward deregulation of economic
activity, as in China in the 1980s, have improved the
competitive environment under which energy deci-
sions are taken and thus have contributed to im-
proved energy efficiency.
37 Donald Hertzmark, "Energy Efficiency and Energy Pricing in Developing Countries," op. cit., footnote 26.
^Samuel Baldwin, Howard Geller, Gautan Dutt, and N.H. Ravindramath, "Improved Woodburnin g Cookstoves: Signs of Success,' AMB10, vol.
14, No. 4-5, 1985, pp. 280-287.
Chapter 2— Energy and Economic Development • 39
Box 2-B — Energy Conservation Initiatives in ASEAN Countries
In Southeast Asia, many governments are adopting and implementing laws to encourage energy conservation
in buildings and industry. Design standards have been enact@or are being considered, in most ASEAN countries.
In Malaysia, the Ministry of Energy, Telecommunications, and Posts has embarked on development of energy
standards for new buildings, with the goal of reducing overall usage by 10 percent by 1991 (5 to 15 percent for
lighting, 5 to 10 percent for air-conditioning, and 15 to 20 percent relating to heat gain through building envelopes).
These standards were widely circulated for review, and were expected to be implemented during 1989. Some energy
audits have been commissioned. 1
In the Philippines, major energy consumers are required by law to have energy management programs, and
large customers must report their consumption to the Office of Energy Affairs (OEA) quarterly. The OEA offers
a wide range of conservation services, including an energy management training program, energy conservation
briefings, industry-specific publications, assistance to the Energy Management Association of the Philippines (a
private-sector group), consulting and audits, efficiency testing, and industrial efficiency monitoring. The Omnibus
Energy Conservation Law mandates the development of standards for energy use in commercial buildings, for
building construction materials, and for designs of commercial and industrial buildings prior to the issuance of
permits for building or for adding equipment such as air-conditioning units.
Singapore encourages conservation through educational programs. Indirect controls imposed by the
government's Building Control Department standardize various design features, such as overhangs and reduced
window area to decrease demands for air conditioning.
In Thailand, the government's Sixth National Economic and Social Development Plan (1987-91) specified
targets for increased efficiency in transportation, industry, and households. Tax reductions and low-interest loans
for energy conservation equipment are available.
1 The 1986 standards instituted in Malaysia areas follows: 1) Buildings whose connected electric service is over 250 kVA are required
to have separate meters for lighting and outlets and for air conditioning systems. 2) Lighting loads are specified for interior spaces, several
building interiors, and roads and grounds in the vicinity of the building. Lighting controls are specified. 3) For air conditioning, dry bulb
temperatures are set at 25 degrees Celsius and relative humidity at 6.percent, plus or minus 5 percent. Automatic setback and shutoff systems
are required.
Energy and The Traditional Sector
Two-thirds of the developing world's population-
some 2.5 billion people-live in rural areas 39 with
low standards of living based largely on low-
resource farming. This type of farming is character-
ized by high labor requirements, low productivity
per hectare and, because of the marginal subsistence,
strong risk aversion. Rural populations have little
access to commercial fuels and technologies and
only limited connection with the modern economy.
Biomass fuels satisfy the heating and cooking needs
of these populations, and muscle power largely
provides for their agricultural, industrial, and trans-
portation energy needs. Although these energy
sources provide crucial energy services at little or no
direct financial cost, biomass fuels, muscle power,
and related traditional technologies generally have
low efficiencies and limited output and productivity
levels (see ch. 3).
In many areas, biomass supplies are diminishing
due to a host of factors, including population growth
and the expansion of agricultural lands, commercial
logging, and fuel wood use (see ch. 5). The poorest
rural people often have limited access to even these
resources and, therefore, must spend longer periods
of time foraging for fuel sources-exacerbating their
already difficult economic position.
Traditional villages are complex, highly intercon-
nected systems that are carefully tuned to their
environment and the harsh realities of surviving on
meager resources." Because the villages are largely
closed systems, changes in any one part affect other
39 WorldBank, World Development Report 1989, op. cit., footnote 1.
@See, for example, M.B. Coughenour et «l., "Energy Extraction and Use in a Nomadic Pastoral Ecosystem," Science, vol. 230, No. 4726, Nov. 8,
1985, pp. 619-625; J.S. Singh, Uma Pandey, and A.K. Tiwari, "Man and Forests: A CentrsHimalayan Case Study," AMBIO, vol. 13, No. 2, 1984,
pp. 80-87; Amulya Kumar and N. Reddy, "An Indian Village Agricultural Ecosystem-Case Study of Ungra Village, Part II: Discussion" Biomass,
vol. 1, 1981, pp. 77-88.
40 • Energy in Developing Countries
elements of village life. Changes in agricultural
practices, for example, change the amount and type
of energy supplies available. In turn, energy sector
developments, such as rural electrification, can have
major impacts on agricultural practice and income
distribution. Making changes in rural systems fre-
quently proves difficult due to the large risks that
changes can pose to populations living on the margin
of subsistence.
The following sections examine four of the major
factors that affect the linkages between energy and
the economic and social development of rural
economies: seasonality; inequities in the distribu-
tion of and access to resources; the role of commer-
cial biomass in the rural economy; and gender issues
in labor. Mechanizing the mundane tasks of rural
life, a process facilitated by the introduction of
modern fuels, could greatly increase the productivity
of rural peoples. To bring about improvements,
however, will require paying close attention to the
numerous related complications, such as seasonal-
ity, the type of task, culture-specific labor roles,
children's labor, and many others.
Seasonality ,
The seasons affect every aspect of rural life: the
availability of food, fuel, and employment; the
incidence of disease; and even the rates of fertility
and mortality .41 Labor requirements for planting are
seasonally peaked to take advantage of limited
rainfall and other favorable growing conditions.
When rains begin, soil bacteria multiply rapidly and
break down the dead plant matter in the soil left by
the dry season; this process releases a large amount
of organic nitrogen in the soil. Crops planted quickly
after the rainfall can take advantage of this nitrogen,
but a short delay leaves weeds as the main benefici-
aries. 42 Labor requirements to harvest crops are
similarly peaked (see figure 2-9). Thus, while there
may be a large labor surplus during most of the year,
labor shortages occur during the critical planting and
harvesting seasons. Studies of African agriculture
indicate that labor is "the major scarce resource in
food production.' ,43
Modern equipment could reduce the high labor
demands during planting and harvesting. Even when
the necessary commercial fuels are available, how-
ever, modern agricultural equipment is often prohib-
itively costly to purchase or rent due to the very short
period in which it can be profitably used.^Rela-
tively low-cost traditional technologies face similar
cost barriers. For example, the average animal-
drawn cart in Ungra, India, is used at just 6 percent
of its annual capacity .45
Draft animal technologies can ease critical sea-
sonal labor shortages to some extent. Draft animals,
however, can only be used productively for little
more than the short growing season, yet these
animals require food year round. Limited uses for
draft animals, coupled with their high food require-
ments, reduce the average draft animal efficiencies
to just a few percent. Because of the limited supplies
of fodder available, farmers often semi-starve draft
animals in order to save fodder for when the animals
need their strength to plow the dry-baked ground or
for other purposes. 46 A shortage of draft animals
may limit crops to just one per year-even in areas
with potential for double cropping. 47
Although agriculture demands very high levels of
labor during the peak seasons, during the remainder
of the year, rural areas experience serious under-
41 Robert Chambers, Richard Longhurst, and Arnold Pacey (eds.), Seasonal Dimensions to Rural Poverty (London: Frances Pinter Publishers, Ltd.,
and1bttawa,NJ: AHanheld,Osmun&Co., 1981); Robert Chambers, "RuralPoverty Unperceived: Problems and Remedies,"WorWDevefo/>menf , vol.
9, 1981, pp. 1-19.
4! Robert Chambers, Richard Longhurst, and Arnold Pacey (eds.), Seasonal Dimensions to Rural Poverty, op. cit., footnote 41, pp. 10-11.
43 Jeanne Koopman Henn, "Feeding ** Cities and Feeding the Peasants: What Role for Africa's Women Farmers?" World Development, vol. 11,
No. 12, 1983, pp. 1043-1055.
^Prabhu Pingali, Yves Bigot, and Hans P. Binswanger, Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa
(Baltimore, MD: Johns Hopkins University Press for the WorldBank, 1987).
45 As a consequence of low utilization rates, villagers preferred lower cost wooden wheels over higher performance pneumatic tires until the depletion
of timber resources caused the price of wooden wheels to rise significantly. At the same time, these price rises led to an active market in second-hand
traditional carts. H.I. Somashekar, N Ji Ravindranath, and Amulya Kumar N. Reddy, Studies on the Ungra Village Agricultural Ecosystem, Part III:
Animal Drawn Carts and Transport (Bangalore, India: AS-no date).
•^HJ. Somashekar, N.H. Ravindranath, and Amulya Kumar N. Reddy, Studies on the Ungra Village Agricultural Ecosystem Part III: Animal Drawn
Carts and Transport, op. cit., footnote 45; Jane Bartlett and David Gibbon,A;ijm<rf Draught Technology: An Annotated Bibliography (London: lilXr,
1984); Peter Mnnzinger, Animal Traction in Africa (Eschborn, West Germany: GTZ, 1982).
47 NK Ravindranath and H.N. Chanakya, -si-^ Based Energy System fOr a South Indian Village, " Biomass, vol. 9, No. 3,1 986, pp. 215-233.
Chapter 2™Energy and Economic Development . 41
Figure 2-9-Seasonal Pattern of Energy Expenditure on Agricultural and Other
Essential Tasks: Adult Farmers, Genieri Village
.„„„ Kcal/day
2000 y,
1600
1200
800
Females
J
H -,s = +_
Female annual
average
Male annual
H
I
400
1
2
I
1 !
Y
1
1
1
Males
1
-I I
I
I
- 7 .
I
I
f - 1
I
1
1 4
Rainy season
~
I
!> .
/ ,
1
I I
I
I
I L
May June July Aug Sept Oct Nov Dec Jan Feb Mar Apr
— — Males Females
SOURCE: Margaret Haswell, Energy for Subsistence (London: MacMillan Press, Ltd., 1981).
employment. In turn, this seasonal unemployment in
rural areas propels a large amount of both seasonal
and permanent migration to urban areas. 48 In Africa
and Asia, where the migrants are mostly men, 49
more of the burden for subsistence crop production
is shifted to the women who stay behind. Migration
to cities increases pressure on forests as well, be-
cause urban dwellers generally purchase their wood
supplies, which are likely to be derived from cutting
whole trees, rather than the gathering of twigs and
branches as is more typical of rural foragers.
The seasons also affect the availability and
usability of renewable energy resources. During the
rainy season, wood is less easily obtained and more
difficult to burn than during the drier months. I n
areas heavily dependent on crop residues for fuel,
shortages at the end of the dry season can force the
use of noxious weeds as substitutes, particularly by
the very poor. 50 The use of biogas is limited by the
availability of dung, which in Ungra, India, varies
for adult cattle from 3.5 kg/day during the 7-month
dry season to 7.4 kg/day during the wet season. 51
Correspondingly, in mountainous areas or elsewhere
with large seasonal temperature variations, fuel
demands can increase significantly during the
winter. In a Kashmir village, for example, wood
^Michael p. Tbdaro, Economic Development in the Third World (New York, NY: Longman, i nc . ; 1977) ; Gerald M. Meier, Leading Issues in
Economic Development, 4th ed,(New York, NY: Oxford University Press, 1984); Scott M. Swinton, Peasant Farming Practices and Off-Farm
Employment in Puebla,Mexico(lti3aca,NY: Cornell University, 1983).
49 Michael P. Todaro, Economic Development in the Third World, op. cit, footnote 48, pp. 192-193. Note that in Latin America more women than
men now migrate.
50 Varun Vidyarthi, "Energy and the Poor in an Indian Village," World Development, vol . 12, No. 8, 1984, pp. 821-836.
51 N.H.RavTndranath and HJ>f.Chanakya,''Biomass Based Energy System for a South Indian Village," op. Cit., footnote47.
42 •Energy in Developing Countries
demands during the winter are four times higher than
during the summer. 52
Inequities in Resource Distribution
and Access
In regions where biomass fuel supplies are
limited-particularly those with dry climates and/or
high population densities-rural people may travel
long distances to collect fuel for domestic use, as
much as 20 miles round trip in some areas under
special conditions. More generally, when wood is
scarce they rely on crop wastes, animal dung, or
other materials as substitutes. Estimates of time
spent in foraging range as high as 200 to 300
person-days per year per household in Nepal. 53
Foraging is also heavy work. In Burkina Faso,
typical headloads weigh 27 kg (60 pounds) .54 In
many regions, women and children shoulder most of
the burden.
Despite these heavy burdens, villagers often
prefer to invest their capital and labor in technolo-
gies for income-producing activities, such as yarn
spinners, rather than in fuel -conserving stoves or
tree-growing efforts. 55 Reasons for this investment
preference include lack of cash income; the ability to
minimize wood use or to switch to alternative fuels
when wood becomes scarce; 56 conflicts over owner-
ship of land or trees; and easy access to common
lands. In addition, villagers often carry out fuelwood
collection in conjunction with other tasks, such as
walking to and from the fields or herding animals. I n
this case, collecting biomass resources may prove
less burdensome than it appears. 57
If wood is scarce, villagers use crop wastes, dung,
'or other less desirable fuels. To the village user, the
immediate value of these fuels outweighs their
potential long-term environmental costs. 58 ln India,
for example, a ton of cow dung applied to the fields
produces an estimated increase in grain production
worth $8 (U.S. dollars), but if burned eliminates the
need for firewood worth $27 in the market. 59 The
diversion of crop residues, previously used as soil
enhancers, to fuel use may lead to a long-term loss
in soil fertility unless offset by increased use of
chemical fertilizers.
Local fuel shortages often have their most serious
impacts on rural landless and/or marginal farmers
with little access to fuel supplies. The poor may also
sometimes be denied access to their traditional fuel
sources when the market value of biomass rises. 60
For example, farm laborers in Haryana, India, are
now sometimes paid in crop residues for fuel rather
than in cash, although previously they had free
access to these agricultural wastes. 61
The Role of Women
Women shoulder the burden of most domestic
tasks, including foraging for fuelwood and cooking.
I n many areas they also perform much of the
subsistence agricultural labor. A 1928 survey of 140
Sub-Saharan ethnic groups found that women "car-
ried a major responsibility for food farming" in 85
percent of the cases, and did all but the initial land
52 Maji<JHussain, Fu e' ^nsumption Patterns in High Altitude Zones of Kashmir andLadakh," Energy Environment Monitor (India), vol. 3, No.
2, September 1987, pp. 57-62.
53j S Singh, UmaPandey.and AX.Tiwari, 11ManandForests . A CentralHiinalayan Case Study," AMBIO, vol. 12, No. 2,1984, pp. 80-87; Kedar
Lai Shrestha, Energy Strategies in Nepal and Technological Options (Nepal: Research Center for Applied Science and Technology, Tribhuvan
University, for the End-Use Oriented Global Energy Workshop, SaiPaulo, Brazil, June 1984). The World Bank Energy Sector Assessment for Nepal
estimated that 16 percent of all labor went for fuelwood and animal fodder collection.
E.Ernest, "Fuel Co njgmptiou Among R ura l Families in Upper Volta, West Africa," paper presented at Eighth World Forestry Conference, Jakarta,
Indonesia, 1978.
55 Varan Vidyarthi, "Energy and the Poor in an Indian Village, " op. cit., footnote 50.
^Phil O'Keefe and Barry Munslow, "Resolving the Irresolvable: TheFuelwood Problem in Eastern and SoutherrAfrica," paper presented at the
ESMAP Eastern and Southern Africa Household EnergyPlanning Seminar, Harare, Zimbabwe, Feb. 1-5, 1988.
57 Irene Tinker, "The Real Rural Energy Crisis: Women's Time," Energy Journal, vol. 8, 1987, pp. 125-146.
^Geoffrey Barnard and Lars Kristoferson, Agricultural Residues as Fuel in the Third World (Washington, DC, and London: Earthscan and
International Institute for Environment and Development Energy Information Program, Technical Report No. 4, 1985).
^G.C. Aggarwal and N.T. Singh, "Energy and Economic Returns From Cattle Dung as Manure and Fuel," Energy, vol.9,No. 1, 1984, pp.87-90;
see also G.C. Aggarwal, "Judicious Use of Dung in the Third World," Energy, vol. 14, No. 6, 1989, pp. 349-352; Eric Eckholm et al., Fuelwood: The
Energy Crisis That Won't Go Away (London: Earthscan, 1984), p. 105; Ken Newcombe, World Bank, Energy Department, "AnEconomic Justification
for Rural Afforestation: The Case of Ethiopia," 1984.
«>Varun Vidyarthi, ' 'Energy and the Poor in an Indian Village," Op. Cit., footnote 50.
61 Centre for Science and Environment, The State of India's Environment 1984-85: The Second Citizen's Report (New Delhi: 1985).
Chapter 2~Energy and Economic Development . 43
clearing in 40 percent of the cases. 62 In contrast,
the Muslim custom of Purdah, for example, tends to
keep women near their homes and away from the
fields in Bangladesh. 63 As women's work often does
not produce any cash revenue, opportunities for
introducing energy- and labor-saving technologies
for women's work are limited. Improving labor
productivity and energy efficiency in rural areas will
thus require special attention to the role of women.
The careless introduction of labor-saving technol-
ogies could increase the burden on women. For
example, the introduction of animal or mechanical
traction for land preparation and planting increases
the area that men can cultivate, but does nothing to
assist women in weeding, harvesting, post-harvest
food preparation, storage, and other tasks. 64
The migration of men to look for urban work
leaves women to fulfill traditional male roles as well
as their own. In Uttar Pradesh, India, the malefemale
ratio in villages is 1:1.4 for the working age group of
15 to 50 years. 65 1 n Kenya, a quarter of rural
households are headed by women— in Botswana, 40
percent. 66 Yet the remittances of the migrants can
make an important contribution to rural household
finances.
Children, too, play an important role in rural
labor, freeing adults to perform more difficult
tasks. 67 In Bangladesh, for example, children begin
performing certain tasks as early as age 4 By age 12,
boys become net producers-producing more than
they consume-and are nearly as efficient in wage
work as men. By age 15, boys have produced more
than their cumulative consumption from birth, and
by 22 they have compensated for their own and one
sibling's cumulative consumption. 68 The major role
of children in farming helps explain high fertility
rates in rural areas.
The Role of Commercial Biomass in
the Rural Economy
While much biomass is used locally, rural areas
are also the source of substantial amounts of
fuel wood (both firewood and charcoal) used in
towns. This trade pumps relatively large amounts
of cash into the rural economy and provides much-
needed employment to rural dwellers during non-
agricultural seasons. To supply Ouagadougou, Bur-
kina Faso, with wood during 1975, for example,
required some 325,000 person-days of labor and
generated over $500,000 in income directly and an
additional $2.5 million in income through transport
and distribution. 70 Such marketing networks can be
quite extensive and complex. 71
In many countries, people in the poorest areas,
where conditions do not permit expansion of crop or
animal production and natural woody vegetation is
the only resource, depend heavily on sales of
62 Jeanne KoopmanHenn, "Feeding the Cities and Feeding the Peasan ts: what Role for Africa's Women Farmers? " op. cit., footnote 43.
63 Mead T. Cain, "The Economic Activities of children in a Village in Bangladesh," Population and Development Review, vol. 3, No. 3, September
1977, pp. 201-227; Gloria L. Scott and MarilynCarr, "The Impact of Technology Choice onRural Women in Bangladesh" WorldBank, Staff'Working
Paper No.731, Washington DC, 1985.
**Peter Munzinger, Animal Traction In Africa, Op. cit., footnote 46\
«J.S. Singh, Uma Pandey, and AJC. Tiwari, "Man and Forests: A Central Himala yan Case Study," op. cit., footnote 40.
^World Bank, Population Growth and Policies in Sub-Saharan Africa (Washington, DC: 1986), p. 39.
Ingrid Palmer aas noted: "Children's labor, especially daughters', is usually more significant tban husbands' in easing a work bottleneck for
women." Ingrid Palmer, "Seasonal Dimensions of Women's Roles," in Robert Chambers, Richard Longhurst, and Arnold Pacey (eds.), Seasonal
Dimensions to Rural Poverty, op. cit., footnote 41.
M Mead T.Cain, "The Economic Activities of Children ina Village in Bangladesh,' 'op. Cit., footnote 63.
'The value of commercialized fuelwood and charcoal exceeds 10 percent of the gross domestic product in countries such as Burkina Faso, Ethiopia,
and Rwanda and exceeds 5 percent in Liberia, Indonesia, Zaire, Mali, and Haiti. Philip Wade and Massimo Palmieri, "What Does Fuelwood Really
Cost?" UNASYLVA, vol. 33, No. 131, 1981, pp. 20-23. George F. Taylor, II, andMoustafa Soumare, "Strategies for Forestry Development in the West
African Sahel: An Overview," Rural Africana, Nos. 23 and 24, Fall 1985 and Winter 1986.
TO J.E.M. Arnold, "WoodEnergy and Rural Communities," Natural Resources Forum, vol.3, 1979, pp. 229-252.
71 AlamBerlrand,''^^^ ,m 8NetworksforForestFuels to Supply Urban Centers in the Sahel, "Rural Africana, Nos. 23 and 24, Fall 1985 and Winter
1986.
44 . Energy in Developing Countries
firewood for their income. "Similarly, when crops
fail, charcoal production 73 -or, as in Bangla-
desh, the cutting of wood from farm hedgerows
grown in part as an economic buffer to be sold before
more valuable livestock and land 74 — provides al-
ternatives for earning cash. In India, for example,
"headloading" (individuals carrying wood to urban
markets on their heads) has become an important
source of income for perhaps 2 to 3 million people .75
The response of rural peoples to fuel shortages
varies widely. Some sell wood to urban markets and
use the lower quality residues themselves. Others
use dung for fuel rather than for fertilizer. In Malawi,
to grow sufficient fuel for household use on the
typical family farm would displace maize worth
perhaps 30 times more; collecting "free" wood
proves much easier. 76 In contrast, aerial surveys of
Kenya have shown that hedgerow planting increases
with population density-demonstrating that villag-
ers respond to the reduced opportunity of collecting
free wood from communal lands by growing their
own. 77
Conclusion
High rates of economic growth will be needed in
developing countries to provide the rapidly growing
population with improved living standards. If pres-
ent trends in energy and economic growth continue
into the future, a sharp increase in commercial
energy consumption in developing countries will be
required. Substantial increases in supplies of bio-
mass fuels will also be needed. This prospect raises
a dilemma. On the one hand, increases in energy
supplies on this scale would severely strain finan-
cial, manpower, and environmental resources. But
on the other, inability to supply needed energy can
frustrate economic and social development.
One way to resolve this dilemma lies in distin-
guishing between the energy that is consumed and
the services derived from it. Technological improve-
ments and other means offer the potential of greatly
improving the efficiency of energy use-providing
more of the energy services needed for development
while consuming less energy. The next chapter
provides an introduction to the services provided by
energy, and how they are currently delivered, with a
view to identifying potentials for improving effi-
ciencies.
72 J.E.M. Arnold, "Wood Energy and Rural Communities," op . c j t . ; footnote 70, Centre for Science and Environment, The State of India's
Environment 1984-85: The Second Citizen 's Report, op. ciL, footnote 61.
^D.O. Hall and PJ. de Groot, "Biomass For Fuel and Food-A Parallel Necessity," draft for Advances in Solar Energy, Karl W. Boer (cd.), vol.
3, Jan. 10, 1986; RafiqulHudaChaudhury, "The Seasonality of Prices and Wages in Bangladesh," in Robert Chambers, RichardLonghurst, and Arnold
Pacey (eds.), Seasonal Dimensions to Rural Poverty, op. cit., footnote 41.
74 Rick j. Van Den Beldt, 'Applying Firewood for Household Energy, " j n M. Nurul Islam, Richard Morse, and M. Hadi Soesastro (eds.), Rural
Energy To Meet Development Needs (Boulder, CO: WestView Press, 1984).
75 Centre for Science and Environment, The State of India's Environment 1984-85: The Second Citizen' sReport, op. cit., footnote 61, p. 189.
76d French, "The Economics of BioEnergy in Developing Countries, "in H. Egneus et al. (eds.), Bioenergy 84, Volume V: Bioenergy in Developing
Countries (Amsterdam: Elsevier, 1985). It is estimated that 90 percent ofall rural households collect all their woo @10 percent purchase some of their
wood at $0.50/m 3 or $0.04/GJ. Urban households buy their wood at a cost of $0.12/GJ. In contrast, plantation-derived fuelwood can cost $1.50 to
$2.00/GJ. A farmer could plant trees, but the loss of 0.4 hectare of farmland reduces maize production by a total of $125 and profit by $30. contrast,
trees produced on 0.4 hectare will be worth $6 in 7 years.
■"P .N. Bradley, N. Chavangi, and A. Van Gelder, "Development Research and Energy Planning in Kenya," AMBIO, vol. 14, Nos. 4-5, 1985, pp.
228-236.
Chapter 3
Energy Services in
Developing Countries
Contents
Page
I introduction and Summary * *, .,.*,„. .. , 47
The Residential/Commercial Sector **. ***. .*, .., ,.. ,. *q ( ** ( +**,***+ 50
oooKing + .. $ ■■■<;■ ++q4 @ 4 51
LiohtinQ * * * ** * * * **p *** **** rr
Space Conditioning, Refrigeration, and Other Appliances *..*..,.,..***., 58
The Industrial Sector * **. *.. *.** .**. +4@t * **a, 62
Process Heat ,,, 63
Mechanical Drive * *** * ** ** ** *y* *** c-?
Barriers to Efficiency I mprovements 71
The Traction (Agricultural) and Transportation Sectors **,.,*,*...**.**..*,. 72
Traction and Agriculture e #. *.*. *.** .. + * . < * c . . + . . 72
Transportation 76
Conclusion * * *** * ** **** * ####hp/gj|: on
Appendix 3-A: Energy Balances for Selected Developing Countries 81
Appendix 3-B: Sources for Table 3-9 88
Box
Box page
■3~/\. 1 l*v v/nC~*v/X r I UW • «&>. *%%%%%*^***%%*% ,j ,..,....••■» fj
Figures
Figure Page
3-L Per-Capita Energy Use in Village Households in Developing Countries 50
3-2. Choice of Cooking Fuel by Income for Five Medium-Sized Towns in Kenya 52
3-3. Representative Efficiencies and Direct Capital Costs for Various Stoves 53
3-4. Direct Energy Use for Cooking in West J ava, I ndonesia 55
3-5. Daily Load Profiles for Cooking Energy, Pondicherryjndia, 1980 55
3-6. Light Output and Efficiency of Various Lighting Technologies 56
3-7. Costs of Various Lighting Technologies 57
3-8. Changes in Capacity and Type of Installed Electric Lighting Per Household
With I ncome Level South Bombay, I ndia 58
3-9. Household Electricity Use for Lighting v. Household Income, in Brazil 59
3-10. Electric Appliance Ownership in Urban Malaysia by Income Group, 1980 60
3-11. Electric Appliance Ownership in Urban J ava, 1988 60
3-12. Refrigerator Ownership in Beijing, China, 1981-87 ....*., ***... 60
3-13. Reduction in the Real Cost of Refrigerators Over Time in the United States 60
3-14. Per-Capita Steel Consumption v. GNP for Various Countries 66
3-15. Per-Capita Cement Production v. GNP for Various Countries ..?, 66
3-16. Efficiency of Electric Motors in the United States, Brazil, and India 71
3-17. Energy Losses in an Example Electric Motor-Driven Pumping System
in the United States 72
3-18. Use of Agricultural Pumpsets in I ndia 1950-90 76
3-19. Passenger and Freight Transport v. GNP for Selected Countries, 1%0-81 77
Tables
Table Page
3-1. Total Delivered Energy by Sector, in Selected Regions of the World, 1985 49
3-2. Delivered Energy Per Capita by Sector in Selected Regions, 1985 49
3-3. Per Capita Energy Use by Service in Selected Countries +. 49
3-4. Principal Cooking Fuels Used by the World Population, 1976 51
3-5. Typical Air Pollution Emissions From Various Cooking Fuels 52
3-6. Residential Nonheating Electricity Intensity in Selected Countries, 1970 and 1986 61
3-7. Kenyan National Energy Use by Fuel, 1980 * 63
3-8. Annual Consumption of Fuelwood and Charcoal in Kenya by Rural Cottage Industries,
(GJ /Capita 64
3-9. Efficiency of Fuel Use in Traditional (Developing Countries)andModern
(Industrial Countries) Commercial and Industrial Operations „..,.*, 65
3-10 Energy Consumption by Chinese Industry, 1980 65
3-11. Average Energy Intensities of Building Materials 67
3-12. Energy Intensities of End Products Using Alternative Building Materials 67
3-13. Average Daily Household Consumption of Water, Gujurat, India 68
3-14. Industrial Electricity End Use in Brazil, 1984 69
3-15. Projected Electricity Consumption in India by Sector and End Use, 1990 70
3-16. Agricultural I ndicators for Selected Countries +. 74
3-17. Passenger Fleet Size and Growth in Selected Countries 78
3-18. Energy Efficiency of Trucks in Selected Countries *.,. 79
3-19. Comparison of Rail Systems in China, India, and the United States 80
3A-1. Brazil: Energy Supplies and Services, 1987, Exajoules and Percent of
National Total 82
3A-2. China: Energy Supplies and Services, 1987, Exajoules and Percent of
National Total 83
3A-3. India: Energy Supplies and Services, 1985, Exajoules and Percent of
National Total 84
3A-4. Kenya: Energy Supplies and Services, 1980, Petajoules and Percent of
National Total 85
3A-5. Taiwan: Energy Supplies and Services, 1987, Petajoules and Percent of
National Total 86
3A-6. United States: Energy Supplies and Services, 1985, Exajoules and Percent of
National Total 87
Chapter 3
Energy Services in Developing Countries
Introduction and Summary
Energy use in the world's developing countries is
increasing rapidly. In 1960, developing countries
consumed just 15 percent of the world's commercial
fuels; by 1985, their share had increased to about 24
percent. 1 Including traditional fuels, the developing
country share of world energy consumption was
about 28 percent in 1986. Developing countries
might consume as much commercial energy as
today's industrial countries by early in the next
century. 'Factors driving this rapid increase in
energy use include population growth, economic
growth, and increasing urbanization (see ch. 2).
Even with this rapid growth, overall per-capita
energy consumption rates in developing countries in
2025 would be just one-fifth that of the United States
in 1987.
The purpose of this chapter is to examine how
energy is used in developing countries. As explained
in chapter 1, the focus of this report is on the services
energy provides rather than the amount of energy
consumed. The reason for this approach is simple:
energy is not used for its own sake, but rather for the
services it makes possible. For example, wood might
be burned to cook food, heat water, warm a house on
a winter evening, heat an industrial boiler, or to
provide other services. 3 Similarly, diesel and gaso-
line are used primarily to provide transportation
services.
There may be many different means of providing
a desired service, each with its own costs and
benefits. Transportation, for example, might be
provided by bicycles, motorcycles, cars, buses, light
rail, or aircraft. The consumer chooses among these
according to such criteria as cost, comfort, conven-
ience, speed, and aesthetics. Within these consumer
constraints, a more efficient car maybe preferable to
an increase in refinery capacity in order to reduce
capital and/or operating costs or because of its
environmental benefits. Thus, in addition to engi-
neering and economics, energy analyses should also
consider social, cultural, and institutional factors.
Such factors are more readily included in a services
framework than in a conventional energy supply
analysis.
The amount of energy consumed in the main end
use markets-residential and commercial, industry,
and transportation-is examined first; then the
major services provided by energy are examined
within each end-use market. In the household sector,
the services examined are cooking, 4 lighting, space
conditioning, and refrigeration; in industry, process
heat and motor drive; in agriculture, irrigation and
traction; and, finally, transportation. These services
are chosen on the basis of their current or likely
future levels of energy consumption or their social
and economic impacts.
Within this services framework, changes in en-
ergy use are traced from traditional rural areas to
their modern urban counterparts. The progression
from the traditional rural to the modern urban
illuminates well the wide range of technologies now
being used in the developing countries and the
dynamics of how energy use can be expected to
change in the future.
Energy use in traditional rural villages reflects a
much different set of considerations from that of the
modern urban economy. First, traditional energy use
is part of a complex and interdependent biological
system, rather than being based on fossil fuels. The
biomass that is used for fuel is part of a system that
provides food for humans, fodder for animals,
construction materials, fiber for ropes, and even
traditional medicines. Similarly, the bullock that
pulls a plow also provides milk, meat, leather, and
dung for fertilizer or fuel.
Second, people in traditional economies carefully
assess their choices and make complex tradeoffs
'Jayant Salhaye, Andre Ghirardi, and Lee Schipper, "Energy Demandin Developing Countries: A Sectoral Analysis of Recent Trends," Annual
Review of Energy, vol. 12, 1987, p. 253.
^.S. Environmental Protection Agency, "Policy Options for Stabilizing Global Climate," vol. 2, February 1989, p. VII-30. Rapidly Changing
World Scenario.
3 In some cases, particularly in the industrialized countries, wood might be burned in afireplace simply for aesthetic reasons.
4 Water heating is also an important residential/commercial energy service that is often similar to cooking in terms of the technologies used. It is not,
however, explicitly considered here.
-47-
48 •Energy in Developing Countries
between the numerous pressures they face in day-to-
day survival, at a level seldom seen in the modern
economy. Gathering fuel, for example, is not free: it
costs time and personal energy that must be balanced
against all the other demands that one faces,
particularly during the agricultural season when
labor demand is at its peak. There are also complex
tradeoffs involved in gaining access to fuels on
common lands or on privately owned land.
Third, although people in rural areas may use
energy inefficiently in comparison to what is possi-
ble with modern commercial technologies, they use
energy rather efficiently and wisely given the
constraints on their resources, technology, and
capitals They have little choice in this if they are to
survive on their meager resources. Rather than
maximizing production, as is done in modern
industrial society, traditional peoples focus on
minimizing risk in the face of the vagaries of drought
and other natural disasters.
The efficiency and productivity of traditional
energy technologies in developing countries can be
significantly improved. To do so effectively, how-
ever, will require an understanding of the complex
linkages of village life. In general, village popula-
tions operate rationally within their framework; 6
change then requires that the framework be changed
through the introduction of external inputs-
financial, managerial, material, and technical. The
lack of success of many development programs can
be attributed in part to a failure to recognize the
rationality of rural lifestyles and the need to address
the overall framework in which villagers operate.
For the developing countries as a whole, the
residential/commercial and industrial sectors consti-
tute the largest end use energy markets, together
accounting for 85 percent of the energy used by final
consumers when traditional fuels are included.
Transportation accounts for the remaining 15 per-
cent. There are, however, considerable differences
among developing nations/The residential/com-
mercial sector accounts for a particularly high share
of energy use in African countries (mostly in the
form of biomass fuels for cooking), while industry's
share is quite low. Transportation accounts for an
exceptionally high share of the total in Latin
America, whereas its share in I ndia and China is low.
Tables 3-1,3-2, and 3-3 provide sectoral and energy
service breakdowns for the developing countries;
figure 3-1 shows per-capita energy use in rural
households as determined by village surveys in
Africa, Asia, and Latin America. Residential cook-
ing and industrial process heat account for almost
two-thirds of all the energy used in the developing
world. About 40 percent of all energy consumed in
providing these services in developing countries, or
well over a quarter of the total energy consumed in
developing countries, is used in India and China.
Cooking is the single largest energy use in many
developing countries. There is a well-established
transition in cooking fuels associated with higher
incomes, improved supply availability, and urbani-
zation. In rural areas, and in poor urban households,
traditional fuels (wood, crop wastes, and dung) are
used in simple stoves. In more affluent households,
people switch to modern stoves and clean, conven-
ient fuels such as kerosene, Liquefied Petroleum
Gases (LPG), and electricity. Because wood stoves
are relatively inefficient, households that use kero-
sene or LPG can consume significantly less energy
for cooking than those using wood and charcoal.
Lighting technologies follow a similar technolog-
ical progression, from candles or light from wood
frees in some rural areas, to kerosene and butane
lamps, to electricity, which is a highly prized energy
service. Electricity use for lighting rises rapidly with
household income.
Relatively little energy is used for residential
space cooling in developing countries. Space cool-
ing is becoming significant in commercial and
government buildings, however, and energy use for
space cooling is likely to grow rapidly in the future.
'Notable examples of studies on this topic include: N.H.Ravindranathetal., «^n Indian Village Agricultural Ecosystem-Case Study of Ungra
Village, Part I: Main Observations,"Bi'o»iass, vol. 1, No. 1, September 1981, pp. 61-76; Amulya Kumar N. Reddy, "An Indian Village Agricultural
Ecosystem — Case Study of Ungra Village, Part II: Discussion" Biomass, vol. 1, No. 1, September 1981, pp. 77-88; MB. Coughenour et al.,
"EnergyExtraction and Use in a Nomadic Pastoral Ecosystem," Science, vol. 230, No. 4726, Nov. 8, 1985, pp. 619-625.
«Ibid.
^nergyuse at the village level is fairly similar in both quantity and source (biomass), and in application (cooking, subsistence agriculture) throughout
the world. Energy use by the economically well off is also reasonably similar throughout the industrial countries as well as among the urban elite in
developing countries. The large differences in energy use between countries are due primarily to the relative numbers of villagers and economically well
off in the population; the form and quantity of energy use by those who are making the transition between these two extremes; and the development path
being followed.
Chapter 3~Energy Services in Developing Countries • 49
Table 3-1 — Total Delivered Energy by Sector, in Selected Regions of the World, 1985 (Exajoules)
Residential/commercial Industry Transport Total
Commercial Traditional Commercial Traditional Commercial Traditional Commercial Traditional Total
Region fuels fuels 1 fuels fuels 1 fuels fuels" fuels fuels" energy
Africa 1.0 4.0 2.0 0.2 1.5 — 4.4 4.1 8.5
Latin America 2.3 2.6 4.1 0.8 3.8 — 10.1 3.4 13.5
India and China 7.3 4.7 13.0 0.2 2.0 — 22.2 4.8 27.1
Other Asia 1.9 3.2 4.0 0.4 1.9 — 7.8 3.6 11.3
United States 16.8 — 16.4 — 18.6 — 51.8 — 51.8
-Not available or not applicable.
a these estimates of traditional fuels are lower than those generally observed in field studies. See figure 3-1, app. 3-A, and ch. 4.
b this is delivered energy and does not include conversion losses.
NOTES: 1 exajoule (10" Joules) equals 0.9478 Quads. The residential and commercial sector also includes others (e.g., public services, etc.) that do not fit
in industry or transport. Traditional fuels such as wood are included under commercial fuels for the United States. These figures do not include
conversion losses (from fuel to electricity, in refineries, etc).
SOURCE: International Energy Agency (IE/I), World Energy Statistics and Balances 1971-1987 (Paris: OECD, 1989); \EA,Energy Balances of OECD
Countries 1970-1985 (Paris: OECD, 1987); and IEA, Energy Balances of Developing Counties 1971-1982 (Paris: OECD, 1984).
Table 3-2 — Delivered Energy Per Capita by Sector in Selected Regions, 1985
(gigajoules) (Includes traditionaHuels)
Region Residential/commercial Industry Transport Total
Africa 11.8 5.2 3.5 20.5
Latin America 12.7 12.5 9.7 34.9
India and China 6.7 7.3 1.1 15.1
Other Asia 7.2 6.2 2.7 16.1
United States 69.8 68.5 77.5 215.8
NOTE: These estimates do not include conversion losses in the energy sector and underestimate the quantity of
traditional fuels used compared to that observed in field studies. See app. 3-A for better estimates of traditional
fuel use and for sectoral energy use including conversion losses.
SOURCE: Derived from table 3-1.
Table 3-3-Per Capita Energy Use by Service in Selected Countries (Gigajoules)
Brazil China India Kenya Taiwan U.S.A.
Residential 6.2 11.7 5.5 16.9 8.9 64.9
cooking 5.3 8.5 5.0 16.4 4.7 3.5
lighting 0.3 0.4 0.5 0.5 0.7 —
appliances 0.6 — 0.05 — 3.1 13.0"
Commercial 1.5 0.7 0.26 0.4 4.2 45.2
cooling 0.4 — 0.13 0.24 1.9 —
lighting 0.5 — 0.05 0.16 0.8 7.2
appliances 0.6 — 0.07 — 1.5
Industrial 19.4 13.8 4.1 4.8 39.2 94.1
process heat 17.5 10.2 2.7 — — 55.8
motor drive 1.6 3.6 1.3 — — 20.4
lighting 0.1 — 0.05 — — —
Transport 13.3 1.2 1.3 2.7 11.5 80.8
road 12.0 0.2 0.8 1.8 10.1 66.7
rail 0.2 0.7 0.4 0.2 0.1 2.0
air 0.7 — 0.1 0.7 0.7 11.3
Agriculture 2.1 1.8 0.6 0.5 2.6 2.5
Total 43.4 27.0 11.7 25.6 67.7 288.0
— Not available or not applicable.
'This is the combined total for applicances and lighting.
NOTE: These estimates include the upstream conversion losses in the energy sector, such as the loss in going from fuel to electricity or crude to refined
petroleum products. This in in contrast to tables 3-1 and 3-2 where energy sector conversion losses were not included.
SOURCE: Adapted from app. 3-A, tables 1 through 6.
50 • Energy in Developing Countries
Figure 3-1 — Per-Capita Energy Use in Village
Households in Developing Countries
Per-capita energy use, GJ/year
3 0-
I I i
I I I Id I I I I
Africa
Latin
America
This figure shows per-capita use of biomass fuels as found in
household energy surveys in Asia, Africa, and Latin America. The
observed energy use depends on such factors as fuel availability,
climate, diet, income, and other factors. The generally higher
levels of energy use in Africa and Latin America reflect, in large
part, the greater availability and accessibility of wood and other
biomass fuels. The sporadic peaks in energy use shown in the
figure are typically for villages in colder, more mountainous
regions.
SOURCES: D.O. Hall, G.W. Barnard, and PA Moss, Biomass for Energy
in the Developing Countries (Oxford: Pergamon Press, 1982,
pp. 212; World Bank, "Bolivia: Issues and Options in the
Energy Sector," UNDP/WB Energy Sector Assessment Pro-
gram, Rpt. 4213-60, April 1983; J.S. Singh, U. Pandey, and
A.K.Tiwari, "Man and Forests: A Central Himalayan Case
Study," AMBIO, vol. 12, No. 2, 1984, pp. 80-87; Issoufou
Boureima and Gilles De Chambre, Rapport sur I'Evaiuation du
Programme Foyers Ameliores (Niamey, Niger: Association
des Femmes du Niger and Church World Service, November
1982).
Also, electric appliances are quickly penetrating the
residential sector. Many of these air conditioners and
appliances, notably refrigerators, have low efficien-
cies. These end uses are having strong impacts on the
electric power infrastructure.
Many commercial and industrial processes re-
quire process heat, ranging from the low-tem-
perature heat provided by biomass used to dry food
in cottage industries to the high-temperature proc-
esses used in the large-scale steel and cement
industries. With some exceptions, the efficiencies of
these processes are typically much lower than those
found in industrialized countries.
Much of the population in developing countries
depend for their mechanical work in both industry
and agriculture on human or animal muscle, with
low efficiencies and power outputs that seriously
limit productivity. The efficiencies of modern diesel
and electric motors are significantly lower in devel-
oping countries than in the industrialized countries
as well.
As in other sectors, there is a transition in
transportation technologies. Walking and use of
domesticated animals are the dominant transport
technologies in poorer and rural areas. The next step
up is bicycles, and then the internal combustion
engine. Transport services in the developing world,
as in the industrial world, are based largely on
highways. In the developing countries, however,
freight rather than passenger traffic is the most
important transport activity in terms of energy
consumption.
The Residential/Commercial Sector 8
Energy use in the residential/commercial sector of
developing countries typically accounts for about 30
percent of commercial energy use and two-thirds or
more of traditional fuel use (see app. 3-A). Cooking
is by far the largest use of fuel in rural areas; in urban
and more developed areas, lighting and appliances
(refrigerators and electric fans, for example) are also
large energy users. Air conditioning is likely to
become important in the future in residences and is
already widely used in commercial, institutional,
and government buildings in developing countries.
The average energy efficiency of the most com-
mon cooking, lighting, and appliance technologies
in use in developing countries today can be im-
proved dramatically/but usually at a significant
additional capital cost to the consumer. Neverthe-
less, the advantages of these more modern technol-
ogies-convenience, comfort, effectiveness-are in-
centive enough for consumers to make the invest-
ment where the technologies and the necessary fuel
supplies are available, affordable, 10 and reasonably
reliable.
Bin this analysis, the residential/commercial sector includes other energy uses such as public buildings not included in the industrial and transportation
sectors.
*Tnis can be accomplished by changing both the mix of technology (e.g., shifting users from low-efficiency wood stoves to high-efficiency LPG
stoves) and by improving the individual technologies themselves (e.g., moving toward higheiefficiency refrigerators).
'"Appropriate fioaocial mechanisms ™ay be needed.
Chapter 3™Energy Services in Developing Countries • 51
Table 3-4-Principal Cooking Fuels Used by the World Population, 1976
Percent of people using fuel
Fossil Dung and
Region energy 1 Fuelwood crop waste
Africa South of Sahara 10 63 27
India 10 47 43
Rest of South Asia 12 46 42
East Asia, developing Pacific 36 41 23
Asia centrally planned economies 22 51 27
Middle East, North Africa 53 17 30
Latin America and Caribbean 71 26 3
North America, OECD Pacific 100
Western Europe 100
Europe, centrally planned economies 100
Total 47 33 20
a Includes electric cooking.
SOURCE: Adapted from David Hughart, Prospects for Traditional and Non-Conventional Energy Sources in
Developing Countries, World Bank staff working paper No. 346,132 pp., July 1979.
Cooking "
The most important single energy service in many
developing countries is cooking. In rural areas of
developing countries, traditional fuels-wood, crop
wastes, and dung-are used for cooking; in many
urban areas, charcoal is also used. More than half of
the world's people depend on these crude fuels for
their cooking and other energy needs (see table
3-4). 12 Higher incomes and reliable fuel supplies
enable people to switch to modern stoves and clean
fuels such as kerosene, LPG, and electricity.
Traditional Fuels
Traditional fuels are predominant in rural areas
because they can be gathered at no financial cost and
used in very simple stoves— as simple as an open
free. At the national level, the use of biomass for fuel
reduces expensive energy imports. These are sub-
stantial benefits.
Use of traditional fuels also exacts substantial
costs. Large amounts of labor are expended to gather
these fuels in rural areas, and a significant portion of
household income is spent for them in poor urban
areas (see ch. 2). Cooking with traditional fuels is
awkward and time-consuming. Unlike modern gas
or electric stoves, stoves that use traditional biomass
fuels must be constantly tended to maintain an
adequate flame. This demands a large share of
women's time in developing countries-averaging
perhaps 3 to 5 hours per day 13 — and interferes with
other activities.
Cooking with traditional fuels is also usually
unpleasant and unhealthy due to the large amount of
noxious smoke emitted (see table 3-5). Measure-
ments of indoor concentrations in homes in develop-
ing countries have found levels of carbon monoxide,
particulate, and hydrocarbons 10 to 100 times
higher than World Health Organization standards.
Cooks can be exposed to as much or more carbon
monoxide, formaldehyde, benzo(a)pyrene, and other
toxins and carcinogens as heavy cigarette smokers. 14
Smoke from cooking stoves is therefore thought to
be a significant factor in ill-health in developing
"Although the discussion here focuses on household cooking, the same considerations apply to commercial and institutional settings.
12 Heafing water for bathing and cleaning and boiling water for drinking are implicitly included in the discussion here, as the technologies used are
often the same for the lower and middle income groups in developing countries, and separation of energy use for these purposes Mifficult.
13 Richard Morse etal., "Organizing Current Information for Rural Energy and Development Planning," M. Nurul Islam, Richard Morse, andM.Hadi
Soesastro (eds.), Rural Energy to MeetDevelopment Needs: Asian Village Approaches (Boulder, CO: Westview Press, 1984), table 7, p. 498.
I4 Kiik R - Smith, Biomass Fuels, Air Pollution, and Health: A Global Review (New York, ny ; Plenum Press, 1987)
52 • Energy in Developing Countries
Table 3-5-Typical Air Pollution Emissions From Various Cooking Fuels
GO
Efficiency Grams per gigajoule of delivered energy*
Fuel (percent) TSP so, NOx HC
Wood (tropical) 15 3,800 250 300 3,200
Cow dung (Hawaiian) 15 10,000 3,200 —
Coal (Indian) 20 280 2,200 460 2,200
Coconut husk 15 17,000 — —
Natural gas 60 0.7 — 13 '7
— Not available or not applicable.
a TSP, total Suspended particulates; SO,, sulfur dioxide; NO„ nitrogen oxides; HC, hydrocarbons; CO, carbon monoxide.
SOURCE: Adapted from Kirk R. Smith, Biomass Fuels, Air Pollution, and Health: A Global Review (New York, NY: Plenum Press, 1987).
34,000
44,000
27,000
54,000
330
countries. The diseases implicated include severe
eye irritation, respiratory diseases, and cancer. 15
Finally, although the expansion of agricultural
and grazing lands and commercial logging are the
most important causes of deforestation globally, the
use of wood for fuel may also contribute to deforest-
ation in some local areas, particularly where the
population density is high and the climate is dry such
as the West African Sahel (see ch. 5).
The Transition to Modern Stoves
and Clean Fuels
People are generally observed to make the transi-
tion to modern, efficient stoves and clean fuels as
soon as they are available and affordable (see figure
3-2). "These technologies are preferred for their
convenience, comfort, cleanliness, ease of opera-
tion, speed, and other attributes.
There is a natural progression in efficiency, cost,
and performance as consumers shift from wood
stoves to charcoal, kerosene, LPG or gas, and
electric stoves (see figure 3-3). Improved wood and
charcoal stoves have also begun to fill a potentially
important niche between traditional wood or char-
coal stoves and modern kerosene or gas stoves.
Cultural factors are often cited as a barrier to the
adoption of improved biomass stoves and fuels.
Figure 3-2-Choice of Cooking Fuel by Income for Five
Medium-Sized Towns in Kenya
100
80, 1'
i
60
40
[
20
Percent of income group using fuel
Low
Wood
Middle
Income group
+ Charcoal
High
+ Kerosene
Gas Electricity
Many households use more than one fuel depending on the
particular food cooked and the supply and cost of fuel. Note the
shift in fuel choice from wood to charcoal and kerosene, and then
from charcoal and kerosene to gas and electricity.
SOURCE: John Soussan, "Fuel Transitions Within Households," Discus-
sion paper No. 35, Walter Elkan et al. (eds.), Transitions
Between Traditional and Commercial Energy in the Third Yforld
(Guildford, Surrey, United Kingdom: Surrey Energy Economics
Center, University of Surrey, January 1987).
Although cultural factors may play a role in choices
of stoves or fuels, it is hardly a dominant one, as
evidenced by the wide variety of stoves and fuels
that have already been adopted across the full range
of class, cultural, and income groups in developing
countries. More typically, the reason that various
stoves have not been adopted by targeted groups in
"Although the limited data available linking human exposure to the smoke from wood fires to lung cancer is still ambiguous (but may indicate
anomalously low cancer rates), there is now evidence of excess lung cancer among cooks using certain types of coal in China. Overall, the World Health
Organization now cites respiratory disease from all causes as the leading cause of mortality in developing countries. See Kirk RSmith, "PAHand the
Household Cook in Developing Countries: The Lung Cancer Anomaly," paper presented at the Symposium oiPolvnuclear Aromatic Hydrocarbons
in the Workplace, International Chemical Congress of Pacific Basin Societies, Honolulu, HI, December 1984, to be published in M. Cooke and AJ.
Dennis (eds.), Polynuclear Aromatic Hydrocarbons: Formation, Metabolism and Measurement (Columbus, OH: Batelle Press); J.L. Mumford et al.,
"Lung Cancer and Indoor Air Pollu tionin Xuan Wei, China," Science, vol. 235, Jan. 9, 1 987, pp. 217-220; H.W. de Koning, K.R. Smith, and J.M. Last,
"Biomass Fuel Combustion and Health," Bulletin of the World Health Organization (EFP/84.64).
ls This transition is complex and not yet well understood. Factors that affect a household's shift to modern stoves and fuels include household income
and fuel-producing assets (land, trees, animals, etc.); reliability of access to modeifuels; relative cost of traditional and modern fuels and stoves; level
of education of the head of household; cooking habits; division of labor and control (finances within the household; and the relative performance of
the stoves and fuels available.
Chapter 3™Energy Services in Developing Countries • 53
Figure 3-3-Representative Efficiencies and Direct Capital Costs for Various Stoves
Stove efficiency (%) CapitaLxost ($)
70 r 80
60
Stove capital cost
Improved
charcoal
• stoves
Kerosene
pressure
• stoves
40
20
Electric
hot plate
Animal
Tradition al
Improved
WOO d
Kerosene
dung In
wood
W IC k
traditional
stoves
stoves
• stoves
LPG
stoves
The range of performance both in the laboratory and In the field is much larger than that suggested by
this figure and Is affected by such factors as the size of the stove and pot, the climate (wind), the quality
of the fuel used, the care with which the stove is operated, the type of cooking done, and many other
factors. The type of material that the pot is made of is also a significant factor: aluminum pots are almost
twice as efficient as traditional clay pots due to their better conduction of heat. Although the efficiency
of improved charcoal stoves is shown as slightly higher than improved wood stoves-the case today for
the simplest uninsulated metal woodstoves-the potential performance of wood stoves is higher than
that for charcoal stoves.
SOURCE: Samuel F. Baldwin, Biomass Stoves: Engineering Design, Development, and Dissemination (Arlington, VA:
VITA, 1986); and OTA estimates.
the developing countries is that they simply have not
worked well. 17
The transition to modern stoves and fuels is often
sharply constrained due to their higher capital costs
(figure 3-3) and uncertainty in the supply of fuel. In
Colombo, Sri Lanka, for example, the cost of
converting to LPG in 1983 was equivalent to 1
month's income for 70 percent of the population and
5 months' income for the poorest 20 percent. 18 Yet
cooking with gas can be the lowest cost alternative
when both capital and operating costs are included.
I n Raipur, I ndia, the cost of cooking with LPG is less
than that for wood for household discount rates of 30
percent or less; yet many households continue to use
wood, presumably because effective household dis-
count rates are higher 19 (the capital cost of gas stoves
was cited as a major reason for the failure to switch
from wood to LPG for cooking).
Because of the high cost of LPG cooking, char-
coal and kerosene are widely used as an intermediate
step in the transition from wood to gas stoves.
Charcoal is very popular in some urban areas. For
example, it is the fuel of choice in urban Kenya (see
figure 3-2) and Senegal-which have a tradition of
charcoal production and use remaining from the
historical Saharan trade caravans. 20 Consumers
17 Samuel F. Baldwin, Biomass Stoves: Engineering Design, Development, and Dissemination (Arlington, 'VA: VITA, 1986); Sam Baldwin et al.,
"Improved Woodburning Cookstoves: Signs of Success," AMB10, vol. 14, No. 4-5, 1985.
18 Gerald Leach, Household Energy in South Asia (New York, AT: Elsevier Applied Science, 1987).
19 J. Dunkerley et al., "Consumption of Fuelwood and Other Household Cooking Fuels in Indian Cities," Energy Policy, January/February 1990, pp.
92-99. "Discount rates" are a measure of the time value that households place on their available cash income.
20 WorldBank, Energy Department, "Review of Household Energy Issues in Africa," draft report, May 1987, p. 3, p. 1.6.
54 • Energy in Developing Countries
prefer charcoal to wood because it gives off less
smoke, 21 blackens pots less, requires little tending of
the fire, and in some areas costs I ess. 22 At the
national level, however, cooking with charcoal
consumes far more forest resources than cooking
directly with wood, due to the low energy efficiency
of converting wood to charcoal -typically just 40 to
60 percent and often much lower. 23
Kerosene is usually the next step up in the pro-
gression of cooking fuels. In many areas, kerosene
prices-often subsidized or freed by the govern-
ment—form a reasonably effective cap on the price
of wood and charcoal. 24 Consumers switch between
these fuels according to price and availability.
LPG or natural gas is often the final step in the
progression in cooking fuels. LPG is widely used by
higher income groups in many urban areas, and
natural gas is widely used where it is available. In
Dhaka, Bangladesh, for example, over 50 percent of
the urban population use natural gas; less than 10
percent use kerosene; and none use charcoal. 25 In
some cases, electricity is also used for cooking by
the highest income groups.
As households make the transition from wood to
modern fuels, overall energy use for cooking can
vary dramatically, depending on the choice of
technology and the situation in which it is used.
Total household energy use for cooking with kero-
sene (see figure 3-4) or with LPG can be signifi-
cantly less 26 than that for wood or charcoal, due to
the higher efficiency of kerosene and gas stoves. 27
Total household energy use for cooking with kero-
sene or LPG is also significantly less than for
cooking with charcoal or (non-hydro) electricity,
due to the low conversion efficiency of wood to
charcoal and of fuel to electricity. 28
The transition to modern stoves and fuels thus
offers users many benefits- reduced time, labor,
and possibly fuel use for cooking, and reduced local
air pollution. Means of lowering capital and operat-
ing costs and ensuring the reliability of supply are
needed if the poor are to gain access to these clean,
high-efficiency technologies. At the national level,
the transition to modern stoves and fuels could
improve the local environment 29 and significantly
reduce biomass energy consumption for cooking; to
realize these benefits, however, could impose a
substantial financial burden on poor nations.
A large-scale transition to LPG would require a
significant investment in both capital equipment and
ongoing fuel costs, optimistically assuming that the
cost of LPG systems would average $10 per capita,
the investment would be roughly 3.5 percent of GNP
and 20 percent of the value added in manufacturing
2l Charcoal stoves can, however, give off hi@ levels of carbon monoxide-a serious health hazard in inadequately ventilated kitchens-but this does
not cause as much obvious discomfort to the user as the smoke from a wood fire.
22 Douglas F . Barnes, World Bank, Household Energy Unit, Industry and Energy Department, ' 'Understanding Fuelwood Prices in Developing
Nations," Oct. 31, 1989, table 1, Conversion to dollars per unit of energy was done using 30 MJ/kg for charcoal, and using 700 kg per cubic meter
multiplied by loMJ/kg for wood with typical moisture contents observed in the market.
^The energy efficiency of the conversion process is variously given as 15 percent in Tanzani^ 24 percent in Kenya with an additional loss of 5 percent
of the charcoal itself during distribution?}) 29 percent in Senegal and Ethiopia, and over 50 percent in Brazil with brick kilns. Advanced retorts are claimed
to be capable of achieving 72 percent energy efficiencies in converting wood to charcoal if there is complete recovery of all the gaseous byproducts.
See E. Uhart, Preliminary Charcoal Survey in Ethiopia, UN. Economic Commission for Africa, FAO Forest Industries Advisory for Africa, Dec.
M75-1 122, 1975,30 pp.; M.J. and MX. Luhanga, Energy Demand Structures in Rural Tanzania, Department of Electrical Engineering, University of
Dar-es-Salaam, Tanzania, 1984; Phil O'Keefe, Paul Raskin, and Steve Bemow, Energy and Development in Kenya: Opportunities and Constraints
(Sweden: Beijen Institute, 1984); G.E.Karch, Carbonization: Final Technical Report of Forest Energy Specialist, UNFAO, SEN/78/002, 1980.; T.S.
Wood, Report on Domestic Energy Usefor Cooking (Energy Assessment Mission, Ethiopia) (Washington, DC: World Bank, 1983), p. 33; FLORASA,
Man-h-fade Forests for Wood and Charcoal in Brazil (Minas Gerais, Brazil: Florestal Acesita, S.A., Belo Horizonte, October 1983), p. 53.
Douglas F, Barnes, "Understanding Fuelwood Prices in Developing Nations," op. cit., footnote 22.
^MJ. Prior, "Fuel Markets in Urban Bangladesh," WorldDevelopment, vol.14.iVo. 7, pp. H5-872.
M Not all the potential efficiency gains of LPG may be realized by the vex-y poor. For example, one-third to one-half of the poor in hillside shanty towns
of Rio de Janeiro own just one LPG bottle. To avoid the risk of running out of gas and having no substitute, many households exchange their gas bottles
before they are completely empty. See Alfredo Behrens, Household Energy Consumption in Rio De Janeiro Shanty Towns (Rio de Janerio, Brazil:
Colegio da America Latina, 1985).
27 In practice however, the savings with LPG Me not quite as large as would be expected from thehighet efficiency and better control of these stoves.
This may be due, in part, to less precise control of the stove; to taking advantage of greater useful energy; and other factors. See Kevin B. Fitzgerald,
Douglas Barnes, and Gordon McGranahan, "Interfuel Substitution and Changes in the Way Households Use Energy: The Case of Cooking and Lighting
Behavior in Urban Java," U.N. Working Paper on Interfuel Substitution Analysis, June 13, 1990.
^Other factors that affect household ener gy use f° r cooking include the size of the household, the diet, and the amount of processed or prepared foods
eaten.
29i( might, however, increase global carbon dioxide emissions.
Chapter 3-Energy Services in Developing Countries .55
Figure 3-4 — Direct Energy Use for Cooking in West
Java, Indonesia
Figure 3-5-Daily Load Profiles for Cooking Energy,
Pondicherry, India, 1980
3 5
30
25
20-
15
10
Per-capita energy use, MJ/day
Lower middle Upper middle
Income group
High
'"Wood ~ Kerosene
This figure compares energy use in households using only wood
with that in households using only kerosene within the same
income class. Households using kerosene consume roughly half
as much energy as households using wood.
SOURCE: M.HadiSoesastro, "Policy Analysis of Rural Household Energy
Needs in West Java," M. Nurul Islam, Richard Morse, and M.
Hadi Soesastro (eds.), Rural Energy to Meet Development
Needs: Asian Village Approaches (Boulder, CO: Westview
Press, 1984).
for the nearly three billion people in the lowest
income countries. 30 The LPG used 31 would be
equivalent to one-fourth of the total commercial
energy consumption today by these countries and
would be a significant fraction of their export
earnings. ''Significant economic growth is needed if
these costs are to be absorbed.
Costs would be even higher if electricity were
used for cooking. Direct capital costs for electric
burners typically approach $100 per household or
more. Moreover, at the national level, the capital
cost of installing generation, transmission, and
distribution equipment to power electric burners is
much greater, perhaps several thousand dollars per
household. 33 If relatively few households are using
electricity for cooking, these high capital costs are
partially offset by the numerous other uses for
Per-capita power demand for cooking, watts
1,200 -
/
800-
600-
400-
200-
,J_
\
- T T "' '- r - - r — T
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Hour of day
This figure illustrates the highly peaked power demand for cooking
energy as measured in a village survey.
SOURCE: C.L.Gupta, K. Usha Rae, and VAVasudevaraju, "Domestic
Energy Consumption in India (Pondicherry Region)," Energy,
vol. 5, pp. 1213-1222.
electric power throughout the day. If a significant
fraction of households switch to electricity for
cooking, however, the highly peaked energy demand
for cooking (see figure 3-5) will overwhelm other
baseload applications, and these costs must increas-
ingly be assigned to cooking alone.
Lighting 34
Lighting accounts for only a small fraction of total
national energy use in both developing and indus-
trial countries. In Kenya, for example, just 1.7 per-
cent of national energy use is for domestic lighting
(app. 3-A). Lighting does, however, account for a
significant fraction of total electricity use, and the
electricity sector is very capital intensive (see ch. 4).
Despite its relatively low energy use, lighting
merits particular attention as it plays a very impor-
soWorldBank, World Development Report, 1989 (New York, NY: Oxford University Press, 1989), tables 1 and 6.
■"Assuming aper-eapitapower rate for cooking with LPG systems of 100 watts. This is comparable to that seen in the United States and about twice
that seen in European countries. It is likely that people in developing counties would continue to eat less processed food, less restaurant food, and
probably more grains and so would continue to use somewhat more fuel than is used in households in the industrialized countries. Energy use rates for
household cooking in different countries are given in K. Krishna Prasad, "Cooking Energy," workshop on end-use focused global energy strategy,
Princeton University, Princeton, NJ, Apr. 21-29, 1982.
32 World Bank, WorldDevelopmentReport, 1989, op. cit., footnote 30, table 5. Kilograms of oil equivalent have been converted to energy at42 MJ/kg.
33 Assuming a P ea ' c P 0WCT d eman< ' f 2 kW. a peak power demand is assumed here rather than an average power as for the LPG case above, because
electric power systems cannot easily store power and must be able to meet peak demands.
^Principal sources for the information in this section are Robert van der Piss and A.B. de Giaaf, World Bank, "A Comparison of Lamps for Domestic
Lighting in Developing Countries," energy series paper No. 6, June 1988; and Robert van der Plas, World Bank, "Domestic Lighting, " Energy Sector
Management and Assessments, Industry and Energy Department, working paper No. WPS8, November 1988.
56 •Energy in Developing Countries
tant social role in domestic life and in commerce and
industry, making activities possible at night or
where natural lighting is inadequate. As rural
incomes increase, or as people move to urban areas
and gain greater access to modern fuels and electric-
ity, lighting services and the energy used to provide
them increase dramatically.
Lighting technologies follow a fairly clear tech-
nological progression in performance, efficiency,
and cost (see figure 3-6). Consumers' choices of
lighting technologies largely follow the same pro-
gression as household incomes increase and as
electricity becomes available.
In traditional rural areas, people are often limited
to the light available from wood fires, frequently
obtained in conjunction with cooking. Kerosene
wick lamps are usually the first step up in the
progression. These may be as simple as a wick in a
jar of kerosene, or as complex as a hurricane lamp
with a glass chimney .35 Glass chimney lamps
generally provide more light and at a higher effi-
ciency than open wick lamps. Glass chimney lamps
also cost slightly more-a few dollars-and use
somewhat more fuel. These additional costs can be
a substantial barrier to their use in rural areas. For
example, a survey of six villages in Bangalore, India,
found that three-fourths of the households used
simple open-wick lamps, and only one-fourth used
lamps with glass chimneys. 36
The light provided by wood fires, candles, or
kerosene wick lamps is sufficient to find one's way,
but is generally inadequate for tasks such as reading
or fine work. Using two lamps doubles the cost, but
does not come close to providing adequate light to
work by. Thus, the poorest households tend to use
just one lamp. Wealthier households may add an
additional lamp or two for other rooms in the house
or move up to a kerosene mantle light; however, the
amount of kerosene used per household does not
generally increase in proportion with income. As a
result, the amount of kerosene used for lighting is
similar (within a factor of two or so) across different
Figure 3-6-Light Output and Efficiency of Various
Lighting Technologies
Flux (lumens)
Device efficiency (lumens/watt)
200
andle
Wick
Mantle Incandes-
Floures
Lighting
technology cent
cent
Flux
lumens)
+ Device efficiency
Includes the candle, kerosene wick lamp, kerosene mantle lamp,
60-watt Incandescent lamp, and 22-watt standard fluorescent
lamp. No value is given for a wood fire, as its light output depends
on size and other factors. The light output of candles and
kerosene lamps are similarly highly variable; the values listed are
representative. Only the efficiency of the device (plus ballast) itself
is considered. System efficiencies-induding refinery losses in
kerosene production and generation, transmission, and distribu-
tion losses for electricity-will be considered in a later report of this
OTA study.
SOURCES: Robert van der Has, World Bank, "Domestic Lighting," Energy
sector Management and Assessments, Industry and Energy
Department, working paper No. WPS 68, November 1988.
Van der Has cites the efficiency of electricity production as 30
percent, but this factor is apparently not taken into account in
the incandescent light efficiency figure of 12 InWV. See, for
example, Samuel Berman, "Energy and Lighting," David
Hafemeister, Henry Kelly, and Barbara Levi (ads.), Energy
Sources: Conservation and Renewable (New York, NY:
American Institute of Physics, 1985). Berman gives the output
of a 100-W incandescent as 1,600 lumens and a 50-W
fluorescent as 3,300 lumens. The efficiencies shown here are
slightly iowereorresponding to the lower, assumed wattage of
the light. See also Terry McGowan, "Energy-Efficient Light-
ing," Thomas B. Johansson, Birgit Bodlund, and Robert H.
Williams (eds.), Electricity: Efficient End-Use and New Gener-
ation Technologies and Their Planning Implications (Lund,
Sweden: Lund University Press, 1989).
income groups and in different regions of the
world. 37
Despite the drawbacks of kerosene wick lamps,
they are a predominant technology in poor rural and
urban areas. Although their light output is low, the
capital and operating costs of kerosene wick lamps
are also low (see figure 3-7). Further, kerosene can
be purchased in small quantities as family finances
33 The light given off by wick lamps depends on a host of factors, including size, condition of the wick (unraveled or uneven Land the amount of soot
deposited on the glass chimney (if present).
^ASTRA, "Rural Energy Consumption Patterns: A Field Study," Bangalore, India, 1981.
37 Gerald Leach and MarciaGowen, World Bank, "Household Energy Handbook," technical paper, No. 67, 1987; Suliana Siwatibau, Rural Energy
In Fiji (Ottawa, Canada: International Development Research Center, 1981); Girja Sharan (cd.), Energy Use in Rural Gujurat (New Delhi: Oxford and
D3H Publishing Co., 1987). The Gujurat study found that one of the few variables affecting kerosene use was the number of rooms per household. Even
this, however, was a relatively weak relationship.
Chapter 3— Energy Services in Developing Countries • 57
Total cost to operate ($/hr)
Figure 3-7-Costs of Various Lighting Technologies
Cost per light output ($/lumen)
0.04
0.02
v installation I
r-l
New installations
1.000' -04
Marginal cost
Marginal cost
Wick Mantle Incandescent Flourescent
Lighting technology
.000' I
Wood
handle Wick Mantle Incandescent Fluorescent
Lighting technology
(A) Direct costs to the consumer of operating various lighting technologies per hour of service. (B) Direct costs to the consumer of various
lighting technologies per unit of light output. The costs shown include only cash expenditures; they do not include labor costs for maintaining
kerosene lamps, etc. The high value for electric lights shows the effect of applying all the grid connection charges to a single light
corresponding to the situation faced by the poor rural household that will initially use but one or two lights. The low value for electric lights
ignores the cost of grid connection charges, corresponding to the marginal cost of adding additional lights after being connected to the grid.
The assumed discount rate is 10 percent. In practice, individuals in both the developing and industrialized world tend to apply much higher
discount rates when making investment decisions in energy-conserving technologies. Rates observed in the United States are typically
in the range of 40 to 80 percent. Similarly, high effective discount rates have been observed in developing countries. If higher effective
discount rates are applied, the higher capital costs of kerosene mantle lamps and, especially, electric grid connections will tend to present
more of a barrier to investment.
SOURCES: Derived from figures 3-6 and 3-7A. See also: Harry Chernoff, "individual Purchase Criteria for Energy-Related Durables: The Misuse of Life Cycle
Cost," Energy Journal, vol. 4, No. 4, October 1983,, pp. 81-86; David French, "The Economies of Renewable Enerav Systems for Developing
Countries," Washington, DC, June 1979.
permit . Kerosene wick lamps are thus well matched
to the reality of rural life in developing countries,
where capital and resources are sharply limited.
Wood fires and/or kerosene wick lamps are the
primary sources of light for more than two billion
people worldwide.
Next in the progression are butane or pressurized
kerosene mantle lamps. These are much like the gas
lamps used for camping in the United States. Mantle
lamps give substantially more light and are more
efficient than wick lamps; they also cost more to
purchase and operate, tend to be hot and noisy, and
can cause considerable glare.
Finally, in contrast to kerosene lamps or other
nonelectric lighting technologies, electric lighting is
clean, relatively safe, easy to operate, efficient, and
provides high-quality light. People in rural areas and
small towns of developing countries place electric
lighting high on their list of desired energy services.
For example, a survey of 320 households in several
villages and small towns of Nigeria found that 90
percent ranked electricity— primarily for lighting—
as their top choice in desired energy services. 38
Even where electric lighting is available, how-
ever, the high cost of connecting to the electric grid
creates a substantial barrier for poor families that use
only one or a few lightbulbs (see figure 3-7), 39 and
this substantially slows penetration. A study in
Gujurat, India, found that 10 years after villages had
gained access to the electric grid, less than a third of
the households had connected; this increased to
about two-thirds after 20 years." Uncertain electric
supply in many developing countries-including
blackouts and brownouts-also tends to discourage
potential users and forces those who have connected
to the grid to simultaneously maintain alternative
kerosene lighting systems.
38 EdwardI. Onyebuchi, "Analysis of Rural Energy Choices in Nigeria," Natural Resources Forum, vol. 12, No. 2, 1988, pp. 181-186.
39pjg Ule 3.7 tends t. understate the barrier that grid connection costs present to people in poor rural areas. The perceived and usually the real costs
to finance connection charges are often much higher in developing countries than the 10 percent discount rate assumed for this figure. Using more realistic
effective discount rates of SO percent, the cost of electric lighting — if the villager could raise the money at all — per operating hour would rise from $0.07
to $0.34, compared to $0.02 for a kerosene mantle light and $0,01 for a kerosene wick lamp. The choice of kerosene wick or mantle lamps is thus logical
given the financial constraints that the poor face.
^Girja Snaran (ed.), Energy Use in Rural Gujurat, Op. cit, footnote 37.
33-718
90
58 • Energy in Developing Countries
Electricity use for lighting rises rapidly with
household income. For example, in South Bombay,
India, rates of household electricity use during the
evening varied from 93 watts for the lowest income
group to 365 watts for the highest income group. 41
The "choice of electric lighting technology also
varies as incomes increase. Low-income households
in South Bombay installed more conventional fluo-
rescent lights-despite their higher capital cost—
and operated them more intensively due to their
lower operating costs. As incomes increased, house-
holds shifted away from the harsh light of conven-
tional fluorescent to the more natural light of
incandescent (see figure 3-8). 42
As incomes increase with economic development,
households begin to buy other appliances-radios,
TVs, fans, refrigerators, and air conditioners. Elec-
tricity use for lighting usually continues to increase,
but it becomes only a small fraction of total
residential electricity use (see figure 3-9). Electricity
use for lighting in the commercial and service
sectors also grows rapidly as the economy expands.
The demand for lighting has also continued to
increase in the industrialized countries over the past
30 years as incomes have increased. Today, the
average rate of lighting use ranges from roughly 20
to 100 million lumen-hours per capita per year
(Mlmhr/cap-yr) in the industrial countries. 43 In
comparison, annual household light production in
South Bombay varies with household income from
about 1 to 3 Mlmhr/cap-yr,. 44 light production in the
commercial sector might double these numbers.
This is equivalent to a per -capita consumption level
that is only 10 to 30 percent of the lowest levels
among industrialized countries.
If lighting services equal to half the minimum
level observed in the industrialized countries-10
Mlmhr/cap-yr-are to be provided in developing
countries, then per -capita demand for lighting elec-
tricity will be about 500 (kWh) kilowatthours per
Figure 3-3-Changes in Capacity and Type of Installed
Electric Lighting Per Household With Income Level in
South Bombay, India
Installed wattage, kW Percent fluorescent
50
RS 0-1000
RS 1000-2500
2500-4000
RS 4000-
Income group
Installed wattage + percent fluorescent
Installed wattage per household and the fraction of installed
wattage that is fluorescent (the remainder is incandescent) is
shown versus household income in rupees. The intensity of use
of this installed wattage varied with the type of lighting and the
household income. The lowest income group used 80 percent of
their installed capacity of fluorescent and 45 percent of their
incandescent during the evening. The highest income group
used just 25 percent of their installed capacity of both fluorescents
and incandescents during the evening.
SOURCE: Aehok Gadgil and Bhaskar Natarajan, "Impact of Socio-
Economic and Architectural Factors on Peak Electricity De-
mand: A Case Study of South Bombay," Energy, vol. 14, No. 4,
1969, pp. 229-236.
year. This is equivalent to an evening power demand
of perhaps 150 watts per capita. 45 If that level of
evening demand occurred at the utility system peak
load, as is typical in developing countries, then the
capital cost to provide electricity for lighting would
be roughly $300 per person. 46
Space Conditioning, Refrigeration, and
Other Appliances
Space Conditioning
Heating residential or commercial buildings will
never bean important energy service in the majority
of developing countries since most have tropical
climates. Space heating will be important in some
41 Calculated from data in Ashok Gadgil and Bhaskar Natarajan, "impact of Socio-Economic and Architectural Factors on Peak Electricity Demand:
A Case Study of South Bombay," Energy, vol. 14, No. 4, 1989, pp. 229-236. The lowest income group uses 80 percent of their installed 71 watts of
fluorescent and 45 percent of their installed 81 watts of incandescent; the highest income group uses 25 percent of their installed 1,460 watts of
fluorescents and incandescent.
42 This particular case contrasts with the more typical situation, as discussed for cooking, where the poor are particularly sensitive to first costs.
"Terry McGowan, "Energy-Efficient Lighting," in Thomas B. Johansson, Birgit Bodlund, and Robert H. Williams (eds.), Electricity: Efficient
En&Vse and New Generation Technologies, and Their Planning Implications (Lund, Sweden: Lund University Press, 1988).
^Calculated from Gadgil and Natarajan, op. cit., footnote 41 .
^This assumes that light output is equally split between incandescent and fluorescent lighting, and that the demand is in the 5 hews of the evening.
^Assuming an installed capital cost for the system of $2,000 perkW of generating capacity.
Chapter 3-Energy Services in Developing Countries . 59
Figure 3-9-Household Electricity Use for Lighting
v. Household Income, in Brazil
Electricity use, Brazil (kWh/household-month)
300-
200-
100-
MWU 0-2
MWU 20
MWU 2-5 MWU 5-10 MWU 10-20
income group
Total electricity + Lighting electricity
This graph shows that electricity use for lighting continues to grow
with income even in a relatively properous developing country
such as Brazil. Lighting electricity is, however, only a small fraction
of total household electricity use in this case. MWU are minimum
wage units.
SOURCE: AshokGadgiland GilbertoDe Martino Jannuzzl, Conservation
Potential of Compact fluorescent Lamps in India and Brazil,
Lawrence Berkeley Laboratory and UniversidadeEstadualde
Campinas (Brazil: June 23, 1989).
areas, however, such as mountainous regions and
high-latitude areas like northern China. Beijing, for
example, has about the same annual average low
temperature as Chicago. Nearly 20 percent of
China's total annual coal consumption and 5 percent
of its annual biomass consumption are used for
space heating (a pp. 3- A). 47
In China, residences rarely have any insulation
and often have large gaps around doors and win-
dews. 48 ! ndoor temperatures in these homes are
controlled not by a thermostat or by comfort
requirements, but by fuel supply-and fuel, though
cheap, is scarce. In Kezuo county, Northeast China,
for example, average indoor temperatures are at the
freezing point during the winter, compared to
average outdoor temperatures of-3'C to-5 ^C with
lows of -25 °C. 49 Additions to coal supply, more
efficient stoves, or better wall insulation would thus
result mainly in comfort improvements but not in
energy savings.
Similarly, although many developing countries
have hot climates, 50 little energy is used at present
for space cooling in developing countries. Tradi-
tional building designs somewhat moderate the
extremes in temperature through natural ventilation
and other techniques that make use of local materials
and do not require additional energy inputs. 51
Increasing urbanization and the use of commercial
building materials, however, have made these tradi-
tional practices less practical and less popular.
Active space ventilation by electric fans has become
popular in many areas where there is reliable electric
service and costs are affordable. For example,
electric fan ownership in Beijing, China, jumped
from 47 percent of households in 1981 to 77 percent
in 1984. 52
Air conditioning in residences is a luxury item
found only in the highest income households in
developing countries (see figure 3-10). 53 1 n contrast,
60 percent of all homes in the United States-nearly
all who need it— have air conditioners. 5 ^ substan-
tial proportion of commercial, institutional, and
government buildings in developing countries are
air conditioned.
Air conditioning systems in developing countries
are also often less efficient than those in industrial-
ized countries. Buildings usually are poorly insu-
lated, with large amounts of air infiltration; and air
conditioners are generally less efficient than those in
the west and are poorly maintained and controlled.
47 VaclavSinil,"China'sEnergy,'' contractor report prepared for the Office of Technology Assessment 1990.
^Robert M.Wirtshafter, "Energy-Conservation Standards for Buildings m China," Energy, vol 13, No. 3,1988, pp.265-274; Robert M.Wirtshafter
and Chang Song-ying, "Energy Conservation in Chinese Housing," Energy Policy, vol. 15, No. 2, pp. 158-168.
@ World Bank, "China: County-Level Rural Energy Assessments: A Joint Study of ESMAP and Chinese Experts,' ' Activity Completion Report No.
101/89, May 1989.
50 All 50 of the world's hottest cities are jn the developing world. The hottest is Djibouti, with an average annual high temper atu re of 1 13 °F. None
of the 50 coldest cities is in the developing world. See V. Showers, World Facts andFigures (New York, NY: John Wiley & Sons, 1 979).
5I LimJee Yuan, "Traditional Housing: a Solution to Hopelessness in the Third World: The Malaysian Example," The Ecologist, vol. 18, No. 1,
1988, pp. 16-23; Mehdi N. Bahadori, "Passive Cooling Systems in Iranian Architecture," Scienttfic American, vol. 238,1978, pp. 144154; RJK.Hill,
Utilization of Solar Energy For an Improved Environment Within Housing For the Humid Tropics (Victoria, Australia: CSIRO, 1974).
J2j Sathaye, a. Ghirardi, and L. Schipper, ' 'Energy Denl and in Developing Countries: A Sectoral Analysis of Recent Trends," Annual Review of
Energy, vol. 12, 1987, pp. 253-281.
53 Jayant Sathaye and Stephen Meyers, "Energy Use in Cities of the Developing Countries," Annual Review of Energy, 1985, vol. 10, pp. 109-133.
^Energy Information A<lministration, Housing Characteristics 1984, DOE/EIA-03 14(84) (Washington, DC: U.S. Gov eminent Printing Office,
October 1986), p. 5.
60 •Energy in Developing Countries
Figure 3-10-Electric Appliance Ownership in Urban
Malaysia by Income Group, 1980
Penetration by income group, percent
Bottom 3% Lower 20%
Refrigerators
Middle 28% Middle 29% Upper 15%
Water heaters
Top 5%
income group
Air conditioners
This figure shows the rapid penetration of refrigerators, air
conditioners, and water heaters as household incomes rise. The
incomes (percentage of households) are in ascending order:
150-299 Malaysian dollars per month (M$/month) (3 percent); M$
300-599 (20 percent); M$ 600-999 (28 percent); M$ 1,000-1,999
(29 percent); M$ 2,000-4,999 (15 percent); and M$ 5,000+ (5
percent).
SOURCE: Jayant Sathaye and Stephen Meyers, "Energy Use in Cities of
the Developing Countries," Annual Review of Energy,vo\. 10,
1965, pp. 109-133.
Figure 3-1 I — Electric Appliance Ownership
in Urban Java, 1988
Penetration by income_group,_percent
Low
Lights
Middle
Income group
+ Iron
'Refrigerator
High
+ B&W/Color TV
+ Water pump
This figure shows the rapid penetration and relative importance
within household purchasing patterns of lights, TVs, irons, fans,
refrigerators, and water pumps. Income groups (share of house-
holds) in ascending order are: less than 75 (Rupees)/month (24
percent), 75-120 Rp/month (22 percent), 121-185 Rp/month (21
percent), 186-295 Rp/month (14 percent), and greater than 295
Rp/month (9 percent).
SOURCE: LeeSchipper and Stephen Meyers, "Improving Appliance
Efficiency in Indonesia," Energy Policy forthcoming.
Figure 3-12 — Refrigerator Ownership in
Beijing, China, 1981-1987
Percentage of households
This figure shows the rapid penetration of refrigerators into the
household sector over just a 6-year period.
SOURCE: Stephen Meyers and Jayant Sathaye, "Electricity Use in the
Developing Countries: Changes Since 1970," Energy, vol. 14,
No. 8, 19S9, pp. 435-441, table 6.
Figure 3-13-Reduction in the Real Cost of
Refrigerators Over Time in the United States
Real cost of refrigerators, 1989=1.00
50 100 150 200 250
Cumulative production volume (millions)
Over the past 40 years, the real price of refrigerators has dropped
by almost a factor of 5. For developing countries, such price
reductions would allow households to invest in refrigerators at a
much earlier point in time than was the case for the United States
and other industrialized countries at a similar level of develop-
ment.
SOURCES: Rick Bahr, Bureau of Labor Statistics, U.S. Department of
Labor, personal communication, July 9, 1990 (CPI/
refrigerators); John Chirichiello, National Science Foundation
SRS Computer Bulletin Board, personal communication, July
6, 1990 (GNP deflator, 1 953-1 9S9); U.S. Department of
Commerce, Bureau of the Census, "Historical Statistics of the
United States: Colonial Times to 1970" (Washington, DO: U.S.
Government Printing Office, 1975), p. E1-12 (GNP deflator,
1935-53).
Chapter 3™Energy Services in Developing Countries . 61
Table 3-6-Residential Nonheating Electricity
Intensity in Selected Countries, 1970 and 1986
(kilowatthours per capita)
Country 1970 1986
India 7 25
Indonesia 8 33
Pakistan 10 5 9
Philippines 3 4 78
Thailand 17 110
Malaysia... 37 184
Mexico 75 190
South Korea 2 5 248
Brazil 9 261
Argentina 210 307
Venezuela 158 422
Taiwan 163 557
Japan — 975
West Germany — 1,210
United States — 3,050
—Not available or not applicable.
SOURCE: Stephen Meyers and Jayant Sathaye, "Electricity Use in the
Developing Countries: Changes Since 1970; Energy, vol. 14,
No. 8, 1989, pp. 435-441.
The potential for increased energy use for space
cooling is very large. The United States now uses
about 1,400 kWh of electricity per person per year
for space cooling. 55 If India used this much electric-
ity per person for space cooling, its total annual
electricity generation would have to increase to
more than five times present levels. 56 The hotter
climate of India could increase these requirements
still more.
Refrigerators and Other Appliances
Electricity-using appliances-refrigerators, tele-
visions, washing machines, etc.— are rapidly pene-
trating the residential sector of developing countries.
Factors contributing to this explosive growth in-
clude urbanization, increasing electrification in rural
areas, economic growth, improved access to appli-
ances, and decreasing real costs of appliances—
which make them affordable to a broader segment of
the population than ever before. Factors limiting
appliance penetration include the lack of electric
service, particularly in rural areas. In Brazil, for
example, 90 percent of urban households but only 24
percent of rural households have electric service. 57
The rapidly increasing use of household appli-
ances in the developing countries places additional
demand on electric power infrastructures that are
typically already short of capacity. Further, much of
the residential demand comes at peak times. A
review of 13 of the largest developing countries for
the period 1970-86 found that the growth rate of
electricity consumption was highest in the residen-
tial sector— averaging 9.9 percent annually, com-
pared to 8.3 percent annual growth in the industrial
sect or. 58 Table 3-6 shows electricity intensity for the
residential sector in selected developing and indus-
trialized countries. Even the most advanced devel-
oping countries use, on average, just a small fraction
of the electricity consumed by Americans. Electric-
ity consumption by the economically well off in
developing countries, however, differs little from
that found in the United States or Europe.
Lights are usually the first appliance installed
when a household gets electric service. Acquisition
of other appliances varies by household income and
region (see figures 3-10 and 3-1 1). In India, fans are
typically among the first appliances acquired, fol-
lowed by televisions and refrigerators. In Brazil,
even relatively poorer, newly electrified households
often have televisions and refrigerators, as these
appliances are comparatively inexpensive and are
available secondhand. 59
Refrigerator ownership is at present quite low in
most developing countries. In China, for example,
less than 1 percent of households have refrigerators,
although refrigerator ownership has been growing
rapidly in Beijing (see figure 3-12). In Brazil, 63
percent of households have refrigerators. 3 In con-
trast, in the United States, 99.7 percent of house-
holds have refrigerators. 01
^Gas Research Institute, Strategic Analysis and Energy Forecasting Division, Baseline Projection Data Book (Washington, DC: Gas Research
Institute, 1988), pp. 37, 120.
s^Ikta Energy Research Institute, TERI Energy Data Directory and Yearbook (TEDDY) 1988 (New Delhi, India: 1989), p. 73.
57 GilbertoDeMartino Jannuzzi, "Residential Energy Demand in Brazil by Income Classes," Energy Policy, vol. 17, No. 3, p. 256.
^Stephen Meyersand Ja ya n ' Saula ye." Electr i c '<yUse in theDevelopingCountries: Changes Since 1970, "Energy, vol. 14, JVo. 8,1989, PP-435-441.
"A.Gadgiland G. De Martin} jannuzzi, Conservation Potential of Compact Fluorescent Lumps in India andBrazil, LBL-27210 (Berkeley, CA:
Lawrence Berkeley Laboratory, July 1989), p. 5.
^Howard S. Geller, "Electricity Conservation in Brazil: Status Report and Analysis," contractor report prepared for theOffiee of Technology
Assessment, March 1990, p. 17.
61 Energy Information Administration Housing Characteristics 1984, Op. Cit., footnote 54, p. 13.
62 • Energy in Developing countries
The refrigerators used in developing countries are
typically half the size of American refrigerators, or
smaller. They are also much less efficient than the
best refrigerators now commercially available (the
average refrigerator used in the United States is
similarly much less efficient than the best available).
In Indonesia, most refrigerators are assembled lo-
cally from imported components and, in general, do
not take advantage of proven energy efficiency
features such as rotary compressors and increased
insulation. 62 The efficiency of Brazilian refrigera-
tors is being improved-electricity consumption by
the average new model was reduced by 13 percent
between 1986 and 1989--but they are unable to
make use of the very efficient motor-compressors
(which Brazil manufactures and exports), as these
units cannot tolerate the voltage fluctuations found
in Brazil. 63
Advances in materials and manufacturing tech-
niques, coupled with a growing secondhand market,
are forcing down the first cost of refrigerators and
other appliances. The real cost of new refrigerators
in the United States, for example, has plummeted b y
a factor of nearly 5 over the past 40 years (see figure
3-13). This trend should make many household
appliances affordable to a much larger share of
developing country populations than was the case
for today's industrialized countries at a comparable
level of development— a generation or more ago. As
a result, energy use could increase significantly
above the historical record in the near to mid-term.
For example, the average new refrigerator in the
United States uses about 1,000 kWh of electricity
per year. 64 If every household in China had a
U.S.-style refrigerator, an additional 200,000 giga-
watthours (GWh) of electricity per year-or the
output of about 50 full-size coal-burning power
plants— would be required, at a cost for the power
plants alone of about $100 billion. 65
The Industrial Sector
The industrial sector typically consumes 40 to 60
percent of total commercial fossil energy in develop-
ing countries (see table 3-l); 66 it also makes heavy
use of traditional biomass fuels-often traded in
commercial markets. The primary energy services
required by industry are process heat and mechanical
drive. These services will be treated generically
here; in a later report of this OTA study, they will be
examined as specific parts of integrated industrial
processes.
Firms in the industrial sector of developing
countries today vary widely in size and sophistica-
tion. Atone end of the spectrum are small traditional
firms that use relatively energy-inefficient and
low-productivity manufacturing technologies. 67 At
the other end are large, modern firms, often with
multinational parent companies, that have world-
class manufacturing capabilities.
Manufacturing operations typically fall into three
broad size categories- household or cottage, small
workshops and factories, and large-scale industry.
Over time, a few smaller companies tend to grow
into large ones as the transport infrastructure im-
proves and incomes rise, increasing the size of
markets and providing economies of scale that turn
the advantage to larger firms. 68
In many developing countries, one-half to three-
quarters of manufacturing employment is in house-
hold-scale establishments, with the remainder di-
vided between medium and large operations. Much
of the employment in the small traditional (and
largely rural) household industries is seasonal labor
available during the nonagriculturally active times
of year. Typically one-fourth to one-third of rural
62Le e Schipper, "Efficient Household Electricity Use in Indonesia," Lawrence Berkeley Laboratory, draft report, January 1989, P- 3, section "
"Conservation Potential."
^Howard S. Geller, Electricity Conservation in Brazil: Status Report and Analysis, Op. Cit., footnote 60, P- 29.
"Howard S. Geller, "Residential Equipment Efficiency," contractor report prepared for the Office of Technology Assessment, May 1988.1990,
NAECA standard.
"Assuming: 5 people per household, 45 percent load factor, no tra j^^fo,, and distr ibution losses, and a capital cost of $2,000 peJcW of installed
capacity.
66 J. Sathaye, A. Ghirardi, L. Schipper, ' 'Energy Demand in Developing Countries: A Sectoral Analysis of Recent Trends," op. cit., footnote 52,
table 5.
67 Although they are small and often use little modern technology or methods, these manufacturing enterprises are not inefficient in SOme respects.
High transport and marketing costs and small market size might greatly raise the cost to larger, modern firms if they should try to enter these small village
markets, making them the higher cost producers.
^Dennis Anderson, World Bank, "SmaUtodusu7 m DevelopmgCountries:SomeIssues,"Staff Working Paper No. 518, 1982.
Chapter 3™Energy Services in Developing Countries . 63
Table 3-7-Kenyan National Energy Use by Fuel, 1980 (percent of total)'
Commercial Biomass fuels
Energy service fuels Wood Charcoal Other Total
Household 2.9 46.3 6.6 2.7 58.5
Cooking/heating 1.0 46.3 6.1 2.7
Lighting 1.7 — — —
Other 0.2 — 0.5 — —
Industry 8.6 14.5 1.0 — 24.1
Large 8.6 5.3 0.3 — —
Informal urban — 0.1 0.6 — —
Informal rural — 9.1 0.1 —
Commerce 0.6 0.5 0.1 — 1.2
Transportation 13.7 — — — 13.7
Agriculture 2.5 — — — 2.5
Total 28.4 61.3 7.6 2.7 100.0
— Not available or not applicable.
a Total national energy consumption .332 million gigajoules; per capita power consumption=658 watts.
SOURCE: Phil O'Keefe, Paul Raskin, and Steve Bernow(e<te.), Energy and Development in Kenya: Opportunities and Constraints (Uddevalla. Sweden: Beijer
Institute and Scandinavian Institute of African Studies, 19&).
nonfarm employment is in manufacturing.@This is biomass (wood and charcoal). Informal rural and
an important source of income and employment for urban industries use little or no commercial fuel, but
the rural and poor urban sectors. 70 they account for about 10 percent of total national
energy use in the form of biomass (see table 3-7).
Process Heat Rural applications include beer brewing, black-
Many commercial and industrial processes re- smithing < crop drying ' and pottery firing (see table
quire heat— ranging from the low-temperature heat
used to dry food by cottage industry to the high- Estimates of the use of biomass energy for
temperature processes used by large industries to industrig| gre simj|ar| hj h ^
produce steel and cement. The efficiencies of these Tobacco cu M ri uses n t y Qf « fue| wood , n
processes are typically much lower than those found ||ocos Nort P y hi | ippineS| M and repreS ents 17 percent
in industrialized countries. rf thfi natiQna| ^ ergy budg £ t jn Mala ^ ;i|n
Traditional Process Heat Technologies lndonesia - J e W*- tile ; a f nd ' ime , industr y con ;
3 sumes roughly 2.5 percent of national energy use.
Biomass is used extensively in both traditional Beer brewing uses 14 percent of the total fuelwood
rural and more modern industry in developing consumed in Ouagadougou, Burkina Faso. "Over-
countries. In Kenya, for example, large industry all, biomass fuels supply up to 40 percent of the
accounts for about 8.6 percent of national energy use industrial energy used in Indonesia, 28 percent in
in the form of commercial fuels, and 5.6 percent of Thailand, 17 percent in Brazil, and similarly large
total national energy consumption in the form of fractions in many other countries. 74
tennis Anderson and Mark Leiserson, "Rural Nonfarm Employment in Developing Countries," Economic DevelopmentandCulturalChange,vo\.
28, No. 2, 1980, p. 245, table A2, cited in Donald W. Jones, Energy Requirements for Rural Development (Oak Ridge, TN: Oak Ridge National
Laboratory, June 1988).
^Enyinna Chuta and S.V. Sethuraman (eds.), Rural Small-Scale Industries and Employment in Africa and Asia (Geneva: International Labor Office,
19s4).
7, E.L. Hyman, "The Demand for Woodfuels by Cottage Industries irthe Province of Uocos Norte, Philippines,"£nerg)>, vol. 9, pp. 1-13, 1984;E.M.
Mnzava, "Villagelndustries vs. Savannah Forests," UNASYIVA, vol. 33, No. 131,1981, pp. 24-29; E.M. Mnzava, "Fuelwood and Charcoal in Africa,"
W. Paley, P. ChartierandD.O. Hall (eds.), Energy fromBiomass (London: Applied Science Publishers, Ltd., 1980); MJ. Mwandosyasnd ML. Luhanga,
Energy DemandStructures in Rural Tanzania (Princeton, NJ: Center for Energy and Environmental Studies, Princeton University, and Dar-Es-S_
Tanzania: Department of Electrical Engineering, University of Dar-Es-Salaam, 1984).
72 WorldBank, Energy Sector Management Assistance Program, "Indonesia, Energy Efficiency Improvement in theBrick, Tile and Lime Industries
on Java," March 1987.
^HeimChauviii,' 'When an African City Runs Out of FueL"CWASrXWt,vol.33,No.l33, 1981, pp. 11-20.
74 JoyDunkerley et al., Energy Strategies for Developing Countries (Baltimore, MD: Johns Hopkins University Press, 1981), P. 265.
64 •Energy in Developing Countries
Table 3-8-Annual Consumption of Fuelwood
and Charcoal in Kenya by Rural Cottage Industries,
GJ/Capita
Industry Fuelwood Charcoal 1
Brewing 1 .07
Construction wood 0.50
Butchery 0.24 0.06
Restaurants 0.17 0.04
Baking 0.13 —
Brick firing 0.06
Blacksmithing — 0.06
Crop drying 0.04
Tobacco curing 0.04
Fish curing 0.02
Total 2.27 0.16
—Not available or not applicable.
a This does not include the losses in converting wood to charcoal.
SOURCE: Phil CKsefe, Paul Raskin, and Steve Bemow(eds.) Enwgy and
Development in Kenya: Opportunities and Constraints (Udde-
valla, Sweden: Beijer Institute and Scandinavian Institute of
African Studies, 19S4).
The efficiency with which these tasks axe done
can be quite low (see table 3-9). On close examina-
tion, however, the performance of traditional biomass-
fueled technologies is often found to be carefully
optimized in terms of efficiency, capital, and labor,
given existing materials and technological con-
straints. An example of this is the traditional brick
kiln in Sudan, which holds as many as 100,000
bricks at a time and gains economies through size
and other design factors. To improve the perform-
ance of these technologies usually requires the input
of modem materials and technologies, including
modern means of measuring efficiencies.
Modern Large-Scale Industry
Modern large-scale industries in developing coun-
tries are modeled after their counterparts in industri-
alized countries, but they are often operated at
significantly lower efficiencies. A few energy-
intensive materials-steel, cement, chemicals (espe-
cially fertilizer), and paper-account for much of the
energy used by industry (see table 3-10). The total
energy used to produce these materials will increase
rapidly as developing countries build their national
infrastructures.
Steel— In the OECD countries, the steel industry
typically consumes about one-fifth of the energy
used in the industrial sector. 75 Developing countries
such as China, India, and Brazil devote a similar
share-18 percent, 23 percent, and 20 percent,
respectively 76 -of industrial commercial energy con-
sumption to steel production. The top 10 producers
account for about 90 percent of the crude steel made
in the developing world; many other developing
countries produce little or no steel.
Per -capita steel consumption increases rapidly as
national infrastructures are built, and then tends to
saturate the market and level off at higher income
levels 77 (see figure 3-14). A similar trend has been
found for a wide variety of materials. 78 Simply put,
there is a limit to the number of steel-intensive cars,
refrigerators, washing machines, buildings, bridges,
pipelines, etc., a person needs. Eventually, con-
sumption levels tend to plateau at replacement
levels. When these wants for basic materials are
fulfilled, people tend to spend incremental income
on higher value-added materials-such as those
with a high-quality finish— or on less material-
intensive but higher value-added consumer goods.
The level of per-capita steel consumption needed
to provide a given service has also been reduced over
time through a variety of technological improve-
ments, including higher weight-to-strength steel
alloys, more efficient motors and engines, better
design, and the substitution of alternative products
such as high-performance plastics. For example, the
tensile strength of steel increased fourfold between
1910 and 1980. 79
Overall steel production has been increasing by a
little over 7 percent per year in the developing
countries, while remaining relatively constant in the
industrialized countries. At current rates, steel pro-
duction by developing countries will overtake that in
the industrialized countries early in the next century.
The energy efficiency of steel production in the
developing countries varies widely. In some cases,
it has significantly lagged that of the industrialized
"Maurice v. Meunier and Oscar de Bruyn Kops, "Energy Efficiency in the Steel Industry With Emphasis on Developing Countries," World Bank
technical paper, No. 22, 1984.
76 Ibid.
'Per-capita steel consumption increases approximately linearly with per-capita income up to several thousand dollars.
78 Robert H.Williams Eric D.Larson, and Marc H. Ross, "Materials,Affluence,andIndiistriaIEnergyUse,''y4wiBa/Rev«eH'o/BnCT'jy,vol. 12,1987,
pp. 99-144.
79 Economic Commission for Europe, Evolution of the Specific Consumption of Steel (New York, NY: United Nations, 1984).
Chapter 3— Energy Services in Developing Countries »65
Table 3-9-Efficiency of Fuel Use In Traditional (Developing Countries) and Modern
(Industrial Countries) Commercial and Industrial Operations
Estimated efficiency of
Estimated efficiency of
traditional technology
modern technology in U.S.
Activity
Location
(percent)
(percent)
Cooking
West Africa
15-19
50-60
Beer brewing
Burkina Faso
15-17
79
Burkina Faso
0.3-0.7
0.6
Tobacco drying
Tanzania
0.5
—
Tea drying
Tanzania
2.9
—
Baking
Sudan
12-19
43
India
16
—
Guatemala
3
Fish smoking
Tanzania
2-3
—
Brick firing
Sudan
8-16
6-11
India
6.4
—
Uganda
5-10
—
Foundry work
Indonesia
3
40
NOTE: — Not applicable or not available.
SOURCE: For complete list of sources, see app. 3-B.
Table 3-10-Energy Consumption by
Chinese industry, 1980
Final energy use
Sector Exajoules Percent
Basic metals (iron and steel) 2.38 25.7
Chemicals (fertilizer) 2.23 24.1
Building materials (cement, brick tile) ... 1.44 15.6
Machine building 0.82 8.8
Textiles 0.64 6.9
Food, beverages, tobacco 0.38 4.1
Pulp and paper 0.25 2.7
Other 1.12 12.1
Total 9.26 100.0
SOURCE: World Bank, China: The Energy Sector(Washington, DC: 1985).
countries. Integrated steel plants in India and China
currently use, on average, 45 to 53 gigajoules (GJ )
per ton of crude steel produced; integrated steel
plants in the United States and J apan use half as
much energy .80 Some developing countries have
made significant strides to reduce energy use in steel
production. The Brazilians, for example, cut energy
consumption from 34 GJ to 27 GJ per ton of crude
steel between 1975 and 1979, 81 and the South
Korean steel industry is among the most efficient in
the world.
Cement — The cement industry typically con-
sumes 2 to 6 percent, and sometimes more, of the
commercial energy used in developing countries.
The use of cement is expected to increase rapidly as
national infrastructures of roads, bridges, buildings,
etc., are built. In general, per-capita consumption of
cement increases approximately linearly with in-
come up to several thousand dollars, and then
saturates and levels off at higher incomes (see figure
3-15). Despite the energy intensity of cement
production, it is one of the least energy-intensive
construction materials when in its final form of
concrete/aggregate (see tables 3-11 and 3-12).
The value of cement is quite low compared to its
weight. Because of this and because the raw
materials for cement— limestone, various clay min-
erals, and silica sand-are widely available, cement
is usually produced relatively near its point of use.
In the United States, the maximum range for truck
shipments of cement is about 300 km. In developing
countries, where the transport infrastructure is less
well developed, economical transport distances are
often less. In China, for example, 150 to 200 km is
the typical limit of transport; if transport over longer
distance is needed, the construction of a new cement
plant in the local area will be considered. 82 Thus, as
a result of inadequate transport infrastructures,
cement plants are often small and relatively ineffi-
cient.
The energy required to produce cement varies
widely with the type of production process, quality
^Maurice Y. Meunier and Oscar de Bruyn Kops, op. cit., footnote 75; Sven Eketoip, "Energy Considerations of Classical and New Iron- and
Steel-Making Technology," Energy, vol. 12, No. 10/11, 1987, pp. 1153-1168.
"Maurice Y. Meunier and Oscar de Bruyn Kops, op. cit., footnote 75.
82 Li Taoping, ' 'Cement Industry in China," Rock Products, February 1985, P- 32.
66 •Energy in Developing Countries
Figure 3-14-Per-Capita Steel Consumption v. GNP for Various Countries
Per capita steel consumption, kilogram
U004-
.
1
9 »m
i
t 1
1 •
100
•
9
• .8
• ,m .
•
■ ■
*
10
s
• m
ST.
1
• m
5 10 15
GNP (thousand 1984 $U.S.)
20
The saturation of the steel market at higher income levels is readily seen in linear or logarithmic plots.
It is shown herein a semi-log plot so as to better display both low-end and high-end data. Each data point
represents a country,
SOURCE: United Nations, Statistical Yearbook 1985/86 (New York, NY: 1988), pp. 550-552, table 130.
Figure 3-15-Per-Capita Cement Production v. GNP for Various Countries
Per-capita cement production, kilogram
10000
1000 1
100
10
■ ■
A
■ i
i .
5 10 15
Per-capita GNP (thousands, 1984 $U.S.)
20
The saturation of the cement market at higher income levels is readily seen in linear or logarithmic plots.
It is shown herein a semi-log plot so as to better display both low-end and high-end data Each data point
represents a country.
SOURCE: United Nations, Statistical Yearbook 1985/86, (New York, NY: 1988), pp. 524-526, table 116.
Chapter 3~Energy Services in Developing Countries • 67
Table 3-1 I — Average Energy Intensities of Building
Materials (megajoules per kilogram)
Material Energy intensity
Concrete aggregate 0.18
Concrete 0.80
Brick and tile 3.7
Cement 5.9
Plate glass 25.0
Steel 28J)
SOURCE: Mogens H. Fog and Kishore L. Nadkarni, World Bank, "Energy
Efficiency and Fuel Substitution in the Cement Industry With
Emphasis on Developing Countries," technical paper No. 17,
1983.
of raw materials, plant management and operating
conditions, and other factors. The performance of
cement plants in developing countries also varies
widely and is difficult to characterize simply. Many
plants approach the efficiency of those in the
industrialized countries, depending on when they
were built and the conditions under which they are
operated. Others show significant inefficiencies-
using 25 to 50 percent more energy than efficient
plants of the same type and with the same quality of
raw materials input. 83
Mechanical Drive
Traditional Drive Power
The productivity of people in many rural and poor
urban areas of developing countries is now limited
by their reliance on human and animal muscle power
for water pumping, grain grinding, agricultural
activities, transportation, and small industry. When
only muscle power is available, many hours can be
spent simply on "enabling" activities, such as
hauling water or grinding grain, rather than on
directly economically productive activities. Produc-
tive activities themselves are sharply limited by the
efficiency and total output of muscle power. If the
productivity of people in rural areas of developing
Table 3-12 — Energy Intensities of End Products
Using Alternative Building Materials
(megajoules per square meter)
Structure
Concrete Steel Asphalt
Brick
Building wall 400
Bridge (per m 2 ) 4,000
Roadway (per m 2 ).... 800
8,000
600
3,000 —
— Not applicable or not available.
SOURCE: Mogens H. Fog and Kishore L. Nadkarni, World Bank, "Energy
Efficiency and Fuel Substitution in the Cement Industry With
Emphasis on Developing Countries," Technical Paper No. 17,
1988.
countries is to be increased, modem motor drive
technologies and supporting infrastructures must be
made available at affordable costs. As these technol-
ogies are adopted, energy use-especially electricity-
will increase rapidly.
A person's power output and energy efficiency
are low. The basal metabolism of a person is about
100 watts; for each unit of work output, an additional
4 to 5 units of food energy must be consumed. 84
Working 8 hours a day at a rate of 50 watts of output,
a person consumes about 15 megajoules (MJ ) of
energy and produces 1.5 MJ of work output, for a
daily (24-hour) average efficiency of 10 percent. 85
Much of the labor expended in developing coun-
tries is not directly productive, but is instead for
"enabling' activities-that is, domestic chores.
Hauling water from the village well can take 0.5 to
3 hours per household each day, with a correspond-
ing energy input in the form of food of 0.3 to 3.0
MJ . 86 The poorest households must often go further
and thus have less available time to haul water,
resulting in much lower water usage even with
greater effort (see table 3-13). Water could instead
be pumped by a motor and piped to the home using
just 3 to 5 percent as much energy .87 For electricity
priced at $0.10 per kWh, the direct energy cost for
83 MogensH.FogandKishoreL.Nadkanii,£ner'^) | £^a'e»icyan<iFue/SH6s*i<H<io« in the Cement Industry Witfi Emphasis on Developing Countries
(Washington, DC: World Bank, 1983), see figure 5-1, p. 39.
^W.Edmundson, Energy Research Group, International Development Research Center, Ottawa, Canada, is There a Vicious Cycle Of Low Food
Energy Intake and Low Human Output?" July 1984(Mimeo); Christopher Hurst, Energy Research Group, International DevelopmemResearch Center,
Ottawa, Canada, "Human and Animal Energy in Transition: The Changing Role of Metabolized Energy in Economic Development" June 19S4
(Mimeo); Roger Revelle, "EnergyUselnRural India," Science, vol. 192, June 4, 1976, pp. % 9-975.
^For a countercase, see G.M.O. Maloiy et al., "Energetic Cost of Carrying Loads: Have African Women Discovered an Economic Way?" Nature,
vol. 319, Feb. 20, 1986, pp. 668-669.
''In a study of Gujurat, India, ttre ^ mti required to fetch water was found to vary from 0.5 hour to more than 3 hours per day, with an energy "s e of
100 to 800 kcal/day. Household washing takes 4.5 to 6.3 hours per week and isas strenuous as hauling water. GirjaSharan (etl.), Energy Use in Rural
Gujurat, op. cit., footnote 37.
87 It is often argued ,i,he social interaction provided by activities such as foraging for fuelwood, hauling water, grain @.id@, and others is an
important element of village life and should not be tampered with naively. One notes, however, that village women spend 10 to 12 hours per day in such
activities. Surely they would not object to such social interaction while having a leisurely eup of tea instead.
68 •Energy in Developing Countries
Table 3-1 3- Average Daily Household Consumption
of Water, Gujurat, India
Consumption
class (liters per day)
Landless 60
Less than 2 ha 126
Less than 2-4 ha 134
Less than 4-10 ha 161
More than 10 ha 256
SOURCE: Girja Sharan (cd.), Energy Use in Rural Gujurat (Hem Delhi:
Oxford and BH Publishing Co., 1987).
the typical 1.5 hours spent hauling water would be
just one-fifth of a penny ($0,002). Thus, lack of
access to capital has significant impacts on labor and
energy use.
Similarly, in Africa, to pound maize or millet by
hand can take 1 to 2 person-hours per day per
household. 88 This requires perhaps 1 MJ of energy
(at 300 watts of input). A typical motor-driven mill
can do the same job in a minute or less, with an
energy expenditure of less than 0.2 MJ --or 0.05
kWh. This is less than one-half of a penny ($0,005)
worth of direct energy (at $0.10 per kWh) for 1 to 2
hours' worth of hard labor. The capital costs in these
cases, of course, are a serious barrier to investment;
but with the time saved, the person might have done
something more productive, such as make handi-
crafts for market.
The advantages of mechanical processing of
grains has led to a rapid transition in many parts of
the world. I n J ava, I ndonesia, for example, the
fraction of rice processed by hand dropped from
perhaps 80 percent to less than 40 percent between
1971 and 1973. 89 This freed many women from the
chore of grinding grain; it also cost many of the
poorest households an important source of income
earned by hand pounding rice for wealthier house-
holds. "The introduction of mechanical rice milling
in Bangladesh in the early 1980's was estimated to
displace an additional 100,000 or more poor women
per year from their traditional part-time employment
at hand pounding rice. For the poorest, landless
women, this represented roughly half of their annual
income and 15 percent of family income. 91
The power output and efficiency of draft animals
are similarly limited in performing typical farm
tasks. A typical 500-kg ox or buffalo has a basal
metabolic rate of about 1,000 watts. 92 Average net
output over a 6-hour working day is typically 250
watts, and the net efficiency while working is 29 to
39 percent, which drops to about 10 percent over the
24-hour working day. A typical draft animal might
work just 40 days per year as many of the jobs
formerly done by draft animals-pumping water,
crushing sugar cane, hauling goods to market-have
already been taken over by modern motor-driven
equipment. At such a low rate of usage, the
efficiency of a draft animal is 2 percent or less on an
annual basis. These efficiencies are raised somewhat
when the value of the animal's dung, milk, meat, and
leather is included. 93
The low power output and efficiency of a draft
animal severely restricts the potential work that can
88 PrabhuPingali, Yves Bigot, and Hans P. Binswanger, Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa
(Baltimore and London: Johns Hopkins University Press for the WorldBank, 1987); Mead T. Cain, "The Economic Activities of Children in a Village
in Bangladesh," Population andDevelopmentReview, vol. 3, No. 3, September 1977, pp. 201-227; A.S.Bhalla, "ChoosingTeehniques: Handpounding
V. Machine-Milling of Rice: An Indian Case," Oxford Economic Papers, vol. 17, No. 1, March 1965, pp. 147-157; Margaret Haswell, Energy for
Subsistence (London: MacMillan Press, Ltd., 1981).
8 9C.PeterTimmer,"^ ice °f TechmquemRiraMilm^ in the Third
World (Baltimore, MD: Johns Hopkins University Press, 1984), pp. 278-288. See also A.S.Bhalla, "Choosing Techniques: Handpounding V.
Machine-Miliing of Rice: An Indian Case," op. cit., footnote 88.
90 WilliamL. Collier et al., "A Comment," in Carl K. Eicher and John M. Staatz(eds.), Agricultural Development in the Third World (Baltimore,
MD: Johns Hopkins University Press, 1984).
"GloriaL. Scott and Marilyn Cair, World Bank, * "The Impact of Technology Choice on Rural Women in Bangladesh: Problems and Opportunities,' '
Staff Working Paper No. 731, 1985.
92 PeterLawrenceandAnthonySmith," AB etterBeastof Burden," New Scientist, Apr. 21, 1988, pp. 49-53. Oxen and buffaloes use 2joules permeter
traveled per kg of body weight (2 J/m/kg). Aanimal weighing 500 kg and walking at 1 m/s will use an extra kW, approximately doubling its resting
metabolic rate. Most agricultural animals move at 0.6 to 1.1 mAnimals use more energy for carrying loads than they do for carrying their own weight,
ranging from 2.6 to 4.2 J/m/kg. See alacA.R. Rae, "Bioenergeti.es of Bullock Power,"Energy, vol. 9, No. 6, 1 984; NJH. Ravindranatn et al., "Anlndian
Village Agricultural Ecosystem — Case Study of Ungra Village. Part I. Main Observations," op. cit, footnote 5; Amulya Kumar N. Reddy, "Anlndian
Village Agricultural Ecosystem — Case Study ofUngra Village. Partll. Discussion" op. cit., footnote5. They estimate theefficiency of aulndianbullock
as 8.7 percent when working full time, or if working just 20 days per year as observed, the bullock would have an overall efficiency of 0.5 percent
93 NJI.P^vindranathandH.N.Chaiiakya,''Biomass Based Energy System fora South Indian Village," Biomass, vol. 9, No. 3, 1986, pp. 215-233.
Draft animal efficiency is 3.5 percent, including nitrogen in manure for fertilizer. Without nitrogen, the efficiency is 2.0 percent.
Chapter 3~Energy Services in Developing Countries • 69
Table 3-14-industrial Electricity End Use in Brazil, 1984
Percent of total Fraction of subsector total for each end use (percent)
Industrial electricity Process Direct Electro-
Industry consumption Motor heat heat chemical Light Other
Nonferrous metals 20.9 32 1 35 32 1
Iron and steel 12.4 1 — 98 — 1 —
Chemicals 11.9 79 5 4 9 3 _
Food and beverage 9.0 6 78 16 — 1 3
Paper and pulp . .' 6.5 87 8 2 3
Mining/pelletization 5.6 50 49 — 1
Textiles 5.3 89 4 1 5 1
Steel alloys 4.8 7 — 92 — 1 —
Ceramics 3.9 65 — 34 1
Cement 2.7 91 — 6 3 1
Other 17.0 76 2 16 5 1
Total" 100.0 49 10 32 ■ 2 ~
—Not available or not applicable.
a Total industrial electricity use was 105 terawatthours.
SOURCE: Howard S. Geller, "Electricity Conservation in Brazil: Status Report and Analysis," contractor report prepared for the Office of Technology
Assessment, March 1990.
be done. To irrigate a l-hectare rice crop, for This led to a rapid transition in the industrialized
example, requires the work output of two bullocks, countries from water- and steam-powered drive to
which in turn require the fodder produced from 2 electric drive in the early 1900's; 97 and the electric-
hectares of crop. 94 By himself, the individual farmer ity intensity of industry continues to increase today
could not, however, pump this much water by hand in industrialized as well as developing countries,
in an entire year.
The efficiency of electric motors is generally
fairly high in the industrialized countries, but can be
Modern Drive Technologies significantly lower in developing countries due to
Electric motor drive consumes an estimated 58 to the use of lower quality materials for construction
68 percent of the electricity used in the United and improper techniques for maintenance, repair,
States, and an even higher percentage in the indus- and rewind. 98 Figure 3-16 compares the efficiency of
trial sector alone. 95 Motor drive is similarly impor- electric motors in Brazil, India, and the United
tant in developing countries (see tables 3-14 and States.
3-15). Electric motors are the workhorses of modern .... „. . . ,. ...
industrial societv Thev run home refri aerators- Higher efflcienc y motors are sometimes readily
drive office air condition^; power industrial pumps', ^le m ™£*™*?* {* J"
fan, and compressors; and keep city water supplies ^S^g^^^g Zu-
9 ' facturer of small motors exports more efficient
The efficiency, convenience, and high degree of models than those sold at home. These high-
control of electric motors provide dramatic effi- efficiency motors 99 can not be used in Brazil due to
ciency and productivity improvements in industry/' the excessive variation in the power line voltage.
**Geoffrey Barnard and Lars Kristofferson, Agricultural Residues as Fuel in the Third World (London: Earthscan, 1985).
95 SamuelF. Baldwin, "Energy-Erficient Electric Motor Drive Systems, "in Electricity: Efficient End-Use and New Generation Technologies and
Their Planning Implications (Lund, Sweden: Lund University Press, 1989).
% SamuelF. Baldwin, "The Materials Revolution and Energy-Efficient Electric Motor Drive Systems," Annual Review ofEnergy, vol. 13,1988, pp.
67-94; WX>. Devine, Jr., "Historical Perspective on Electrification in Manufacturing," S. Schunand S. Sonenbhim (eds.), Electricity Use: Productive
Efficiency and Economic Growth (Palo Alto, CA: Electric Power Research Institute, 1986).
^Samuel F. Baldwin, "The Materials Revolution and Energy Efficient Motor Drive Systems, "Annua/ Review of Energy, vol.13, 1988, p. 67-94.
^Samuel F.Baldwin and EmileFinlay, Princeton University, Center for Energy and Environmental Studies, "Energy-Efficient Electric Motor Drive
Systems: A Field Study of the Jamaican Sugar Industry," working paper, No. 94, February 1988. In particular, when motors are rewound they are
sometimes simply put onanopenfire to burn the insulation off the windings rather than in temperature-controlled ovens, This can damage the insulation
between the core laminations and lead to greater losses.
"The efficiency of these motors is equivalent to the standard efficiency in the industrialized countries.
3
to
Table 3-15-Projected Electricity Consumption in India by Sector and End Use,
(percent of total national electricity use)
1990
Industrial process Space conditioning
Motor Process Cooling/ Appliances
Sector Total 1 drive Electrolysis heat Lighting ventilation Heating Refrigeration Other Miscellaneous
Residential 13.0 — — — 4.2 3.5 — 1.5 1.0 2.9
Urban 10.4 — — — 2.9 2.9 — 1.2 1.0 2.4
Rural 2.6 1.3 0.5 — 0.3 — 0.5
Commercial 11.2 — — — 4.8 1.6 1.5 0.4 0.8 2.1
Agriculture 18.4 18.4 ___— — — — —
Industrial 54.8 33.4 10.8 5.5 5.1 — — _ _ _
Primary metals" 17.2 6.4 6.9 3.0 0.9 — — — — —
Chemicals 13.8 8.8 3.6 0.1 1.3 — — — — —
Textiles 10.2 7.8 — 0.4 2.1 — — — — —
Coal, cement 6.8 5.8 — 0.5 0.4 — — — — —
Secondary metals c 3.4 1.5 0.2 1.4 0.2 — — — — —
Paper 3.4 3.0 — 0.1 0.3 — — — — —
Railway traction 2.6 2.6 — — — — — — — —
Total 100.0 54.4 10.8 5.5 14.5 5.1 1.5 1.9 1.8 5.0
Motor drive 61 .4 54.4 — — — 5.1 — 1 .9 —
—Not available or nonapplicable.
a Total national consumption is projected to be 249.1 terawatthours in 1990.
b Aluminum, nonferrous, iron, and steel.
C Iron andsteel.
SOURCE: Ahmad Faruqui, Greg Wilder, and Susan Shaffer, "Application of Demand-Side Management (DSM)To Relieve Electricity Shortages in India," contractor report prepared for the Office
of Technology Assessment, April 1990.
-i
Chapter 3— Energy Services in Developing Countries . 71
Figure 3-16-Efficiency of Electric Motors in the United States, Brazil, and India
Motor efficiency, percent
95
90
85
80
75
70
65
4*=
H — I I I I
10
Electric motor size, HP
Brazil standard eff. + Brazil high eff.
- India field test USA high eff.
~!~- 1 — I
100
India manufacturer
This figure shows the efficiency for motors in Brazil, India, and (high-efficiency only) the United States.
Note the large difference in motor efficiency as measured in field tests and as cited by manufacturers
in India.
SOURCES: United States: John C. Andreas, Energy-Efficient Electric Motors (New York, NY: MarcelDekker, 1982).
Brazil: Howard S. Geller, "ElectricityConservation in Brazil: Status Report and Analysis," contractor report
prepared for the Office of Technology Assessment, March 1990; India: S. Anand, and V.S.Kothari,
Characterization of Electric Motors in Industry and Energy Conservation Potential in India (New Delhi,
India: Tata Energy Research Institute, no date).
This firm has also developed motors with efficien-
cies comparable to the highest performance motors
in industrialized countries. 100
Although the efficiency of electric motors them-
selves can be quite high, the efficiency of the overall
system is generally low. For example, the conver-
sion of coal to electricity typically results in the loss
of two-thirds of the input coal energy. There are
additional losses throughout the system, with the
resulting net output as low as 5 percent of the input
energy (see figure 3-17). Significant energy savings
are possible through the use of better technologies
and better control strategies throughout the system.
Barriers to Efficiency Improvements
A number of factors limit the efficiency, produc-
tivity, and performance of industrial operations in
developing countries: plants that are too small to be
efficient; technologies that are of low quality and
often obsolete; raw materials that are of low quality;
inadequate national infrastructures; lack of foreign
exchange to purchase critical components not avail-
able locally; and a lack of skilled technicians,
engineers, and managers.
The average U.S. paper mill, for example, has an
annual capacity of 100,000 tons, whereas in Latin
America, Africa, and Asia (except J apan), the
average capacities are 18,000,9,000, and 5,000 tons,
respectively. These smaller scales can lead to
significant inefficiencies. Studies indicate that a
paper mill with an annual capacity of 30,000 tons
can consume from 30 percent to as much as 100
percent more energy/steam respectively per unit
output than a mill with a capacity of 150,000 tons. 101
In addition, a variety of energy-conserving technolo-
gies, such as waste heat recovery systems and
100 Howard S. Geller, "Electricity Conservation in Brazil: Status Report and Analysis,' ' contractor report prepared for theOffice of Technology
Assessment, March 1990.
101 Andrcw J> Ewing, "Energy Efficiency in the Pulp and Paper Industry with Emphasis on Developing Countries," WorldBank technical paper, No.
34, Washington, DC, 1985, p. 45.
72 •Energy in Developing Countries
Figure 3-17 — Energy Losses in an Example Electric
Motor-Driven Pumping System in the United States
Fraction of input coal energy remaining
40-
Generation (33%)
20-
Transmission and distribution (90%)
Motor (91%)
Shaft coupling (98%)
Pump 77% Throttle valve
u
|piping
-I 1 7 — 77 —
2 3 4 5 I
Point along the pumping system
7
This figure shows the useful energy remaining at each stage of a
pumping system. The values in parentheses are the efficiencies of
the particular device at each stage.
SOURCE: Samuel F. Baldwin, Energy-Efficient Electric Motor Drive Sys-
tems, in Electricity: Efficient End-Use and New Generation
Technologies and Their Planning Implications (Lund, Sweden:
Lund University Press, 1989).
cogeneration systems, become financially less at-
tractive or even uneconomical at smaller scales.
The raw materials available to industries in
developing countries are often of low quality. For
example, coal resources in India are poor, providing
blast-furnace coke with an ash content that typically
ranges from 21 to 27 percent. This lowers the energy
efficiency of the steel making process as well as
potentially interfering with steel production. 102
Inadequate national infrastructures also reduce
efficiency and productivity. Frequent electric power
brownouts or blackouts are particularly damaging.
In Ghana, for example, the GIHOC Brick and Tile
Co. had 152 hours of electricity outages in 1986.
When an outage occurs, the fuel oil feed to the kiln
burners is cut off and the fire must be stoked with
wood. This is a haphazard process and significantly
reduces the quality of the fried bricks. 103
The lack of foreign exchange to buy spare parts
can also be a serious handicap. This has been an
important factor in the decline of the Tanzanian
cement industry, which operated at just 22 percent of
rated capacity in 1984. 1M
Assistance may be useful at several levels. The
efficiency and productivity of traditional rural in-
dustries might be significantly increased in a cost-
effective reamer with the introduction of a limited
set of modern technologies and management tools.
To do this, however, is extremely difficult due to the
small and scattered nature of traditional rural indus-
tries and the large extension effort needed to reach
it. Large industry in developing countries has many
of the same needs-technical, managerial, and
financial assistance-but can be reached more
readily. International aid agencies and a few non-
governmental organizations are providing such as-
sistance to the extent that their funds allow.
Dramatic improvements in the energy efficiency
and productivity of basic materials processing tech-
nologies are also still possible-even beyond the
levels currently achieved in the industrialized coun-
tries. Research is going on in this area, but much
more could be done. Much of the current research is
focused on higher value-added specialty materials
and high-grade finishing rather than on primary
processing. 105 Developing countries, however, have
little capital to undertake the research needed to
improve their industrial processes, and few interna-
tional agencies support research of this kind.
The Traction (Agricultural) and
Transportation Sectors
Traction and Agriculture
Agriculture entails a series of operations: soil
preparation, sowing, weeding, harvesting, and post-
harvest storage. In traditional agriculture, these
operations are performed by manual labor with, in
some cases, assistance from animals. As agriculture
increases in scale and is commercialized, many of
these operations are done by machines.
Agriculture is by far the largest employer and
goods-producing economic sector in the poorer
developing countries. In sub-Saharan Africa, for
example, 75 percent of the work force is engaged in
102 Energyand Environmental Analysis, Inc., "Conserving Process Heat in Primary Industries of India and China," contractor report prepared for the
Office of Technology Assessment, April 1990.
103 UNDP/World Bank Energy Sector Management Assistance Program, "Ghana: Energy Rationalization in the Industrial Sector," June 1988.
m DP.Stewart,andB.Mutegi/'StrategiesforMeetingTaa^ November 1989, pp. 294-302.
103 Robert H. Williams, Princeton University, personal communication Feb. 1, 1989.
Chapter 3~Energy Services in Developing Countries • 73
agriculture, compared to just 2 percent in the United
States. Agriculture also provides a significant frac-
tion of GDP in developing countries-one-third of
GDP for the nearly 3 billion people in low-income
countries 106 (see table 3-16).
Agriculture in the lowest income countries is
largely by small, family farms using human and
animal power and organic fertilizer with little access
to or knowledge of modern inputs such as chemical
fertilizers, hybrid seeds, or mechanical drive. Ethio-
pians, for example, use on average just 4 kg of
chemical fertilizer per hectare of cropland, while the
English use 368 (see table 3-16). Low soil fertility
and inadequate or irregular rainfall sharply limit the
productivity of low-input farms in developing coun-
tries.
There is a general trend toward larger farms,
greater mechanization, and greater use of commer-
cial inputs in many developing countries, resulting
in greater productivity but at the cost of greater
direct and indirect energy inputs. India, for example,
nearly doubled its irrigated area between 1950 and
1984 in order to reduce its vulnerability to poor
monsoons. 107 1 ncreased irrigation and use of high-
yield variety crops have contributed to increases in
both absolute and per-capita agricultural produc-
tion. 108
China has similarly moved toward greater mecha-
nization and use of modern inputs. Agriculture in
China is sharply constrained by land availability-
only about 10 percent of the land can support
crops-yet per-capita production increased by 18
percent from 1979 to 1983 with little increase in
cultivated area. 109 Improved water control and
distribution, increased use of tractors and fertilizers,
and the adoption of new crop varieties contributed to
this achievement. 110
Traditional Shifting Agriculture
Traditional shifting agriculture begins with forest-
fallow systems, in which plots of forest land are
cleared and cultivated for a few years and then left
fallow for 20 years or more. Clearing by fire requires
little labor, and stumps are left for rapid regrowth
during fallow. Because the ground underneath tree
cover is soft, no further labor is required before
sowing, and because the forest cover has long
suppressed weeds, few seeds remain and little
weeding is needed. Such burning does, however,
effectively lead to very large agricultural energy
intensities due to the large amount of forest cover
that is burned off."
With increasing population density, the fallow
period becomes shorter. As a result, regrowth during
the fallow period is reduced to bush, and finally to
grass. Since fire does not kill roots, extensive hoeing
and weeding become necessary. Inputs of organic
fertilizer are needed to maintain soil fertility, and
there is a shift from simple addition of organic
material to more complex comporting and manuring
techniques. Further increases in population lead to
annual cultivation and eventually multiple cropping.
(In the humid tropics, however, soils tend to be poor
and easily eroded and leached, and the potential for
continuous cultivation is limited. 112 ) As the need for
hoeing and weeding increases it becomes advanta-
geous to go to the extra effort of destumping the land
and obtaining, training, and maintaining animals or
mechanical agricultural technology. 113
There are a number of potential advantages
associated with the use of animal or mechanical
traction for agriculture. Properly done, tillage im-
proves the condition of the soil for crop growth-
increasing porosity, aeration, root penetration, and
water infiltration while reducing evaporation. Ex-
106 World Bank, World Development Report 1990 (New York, NY: Oxford University Press, 1990).
lovrjtfa Energy Research Institute, TERI Energy Data Directory and Yearbook (TEDDY) 1988, Op. Cit., footnote 56, P. 128.
lOSFbod and Agriculture Organization of the United Nations, "The State of Foodand Agriculture, 1984," Rome, Italy, 1985, p. 137.
10 »Ibid.
HOC. Howe, China's Economy (New York, NY: Basic Books, Inc., 1978).
1,1 A. Terry Rambo, "Why Shifting Cultivators Keep Shifting: Understanding Farmer Decision-Making in Traditional Agroforestry Systems,"
Community Forestry: Some Aspects, UNDP THA/8 1/004 (Bangkok, Thailad: Environment and Policy Institute, East- West Center, Honolulu, and
UNFAO Regional Office for Asia and the Pacific, 1984).
1 12 Prabhu Pingali, Yy es Bigot and Hans P. Binswanger, Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa,
op. cit., footnote 88.
1 13 Prabtai Pingali et & ibid. The exact transition point from hand to animal and then tractor technology will, of course, depend onnumero us factors,
including how difficult the soil is to work; the value of milk, me@ hides, and other services provided by cattle; the use of manure; the cost of training
and maintaining animals; the cost of destumping and otherwise preparing fields and weeding; the length of time that animalscan be used; the risk of
disease such as trypanosomiasis (transmitted by the tsetse fly); and many others.
a
t
I
■f
I'
s
a
Table 3-16-Agricultural Indicators for Selected Countries
GNP/Cap Agricultural GDP as Agricultural employment as Percent of farms larger Fertilizer Crop yields (kg/ha)
Country (1987) percent of total GDP percent of total employment than 5 ha (kg/ha) Cereal Roots/tubers
Ethiopia 130 45 — 4 4 1,081 2,827
Zambia 250 13 — — 14 1,747 3,687
India 300 30 61 9 43 1,590 14,268
China 290 26 57 — 176 3,891 15,614
Brazil 2,020 10 36 63 35 1,719 12,072
UK ...0 10,420 2 2 83 368 6,081 36,072
U.S.A 18,530 2 2 9Q 101 4,618 31,215
— Not available or not applicable.
SOURCES: World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 19S9).
Food and Agriculture Organization of the United Nations, TheState of Food and Agriculture 1984 (Rome: FAO, 1965), pp. 163-165, 17S-160.
Tata Energy Research Institute, TERI Energy Data Directory and Yearbook 1988 (New Delhi, India: Tata Energy Research Institute, 1969), p. 123.
World Resources Institute, Wtoritf Resources 7988-89 (New York, N/: Oxford University Press, 19S6), pp. 272-277.
Chapter 3~Energy Services in Developing Countries . 75
Box 3-A — The One-Ox Plow'
Many farmers in developing countries are unable to support the two draft animals needed to pull a traditional
plow. Although half of the households in Bangladesh keep cattle, only a quarter have two or more. In Ethiopia, only
a third of the farmers own two draft animals-and many of these are lost in the periodic droughts. At peak cultivation
times, these farmers must then rent or borrow a second animal and maybe delayed in planting their crops, which
depend critically on catching the sparse and irregular rains in a timely manner-bath for making maximum use of
the nitrogen released with the first rains (seech. 2) and for reaching maturity with the last rains.
Researchers at the International Livestock Center for Africa (ILCA) in Addis Ababa, Ethiopia, responded to
this situation by redesigning the traditional double yoke for a single ox: experiments showed that one ox could pull
with 70 percent of the force of two.
I n the field, however, Ethiopian farmers quickly converted the one-ox plows back into the traditional two-ox
form.0n examination, researchers found that the traditional two-ox form had a number of advantages. Farm oxen
were not as well ted nor as strong as those which had been tested at the I LCA headquarters and could not pull as
hard; and two oxen were able to steady each other when one stumbled. The rigid coupling of the traditional two-ox
plow also enabled the farmer to steer the oxen and to shift some of the weight of the plow to the oxen duringatum.
in contrast, the single ox yoke used a flexible rope harness which reduced the farmer's ability to steer the animals
and forced himtocarrythefull weight of the plow when turning. The one ox plow also had a skid to regulate the
depth of the cut: it broke easily but could not be repaired by the farmers themselves.
Further, where the quality of the feed isverypoor-acommonsituation in many tropical areas-working
animals are unable to compensate for their energy expenditure by eating more and consequently lose weight. A
working animal also has a 10 percent higher basal metabolic rate than a nonworking animal-requiring more food
just for maintenance. I n this case, it may be better to use two oxen to do "what little they can without losing too
much weight rather than to have one ox which soon becomes exhausted beyond recovery. "
As one researcher at the ILCA noted, "It might have occurred to us that if Ethiopian farmers hadn't invented
something as simple as the one-ox plow in 3,000 years of agriculture, they probably had reasons. "
Some have similarly thought that the same animal might be used to provide both labor and milk. Experiments
in Costa Rica showed that cows could, in fact, provide both— if fed adequately. Tropical pastures, however, are not
adequate. To provide the animal a sufficient diet for such a high rate of energy expenditure required concentrated
feed supplements such as grain. This could create a direct conflict over food between draft cows and people in many
parts of the world.
JDebora MacKenzie, "Ethiopia's Hand to the Plough," New Scientist, Oct. 1, 1987, pp. 52-55; Peter Lawrence and Anthony Smith, "A
Better Beast of Burden," New Scientist, Apr. 21,1988, pp. 49-53; A.K.M. Abdul Quader and K. Ikhtyar Omar, Commonwealth Science Council,
"Resources and Energy Potentials in Rural Bangladesh," technical publication series No. 191, London, 1986.
peri merits show that yields can be increased by
plowi ng. w ' n practice, however, little increase is
observed as farmers tend instead to focus on
increasing cultivated area 115 or on saving labor,
rather than improving the quality of their tillage. In
West Africa, the soils are so hard they often cannot
be plowed (without damage to equipment) until the
rains begin, but then any delay reduces the available
growing time and risks a shortage of water when
plants reach maturity. 116
Peasant farmers have responded to their often
difficult circumstances in varied ways— both logical
(see box 3-A) and frequently ingenious. For exam-
pie, around 1925-1930, animal traction began to be
used in northwestern and central Senegal and
northern Nigeria for peanut cultivation. The light,
sandy soils of Senegal do not require plowing, and
as the growing season is so short, rapid planting of
peanuts while the soil is moist is essential. Conse-
quently, seeders are used by the peasants so that
larger areas can be cultivated within the available
n4 PrabhuPingali, Yves Bigot, and Hans P.Binswanger, A gricultural Mechanization and the Evolution of Farming systems in Sub-Saharan Africa,
op. cit., footnote 88; Peter Munzinger, Animal Traction in Africa (Eschborn, West Germany: GTZ, 1987), p. 279.
115 In Senegal the av mg ,„ |1 ansi„„ of agricultural area by the introduction of draft animals to smallholders is 100 to 160percent; in Mali the average
expansion is 150 to 200 percent. Peter Munzinger, Animal Traction in Africa, op. cit, footnote 114, p. 287.
lls PrabhnPingalietai., op. Cit., footnote 88.
76 •Energy in Developing Countries
time. Horses are used instead of oxen, since the
greater power of oxen is not needed (there is no
plowing) and horses are faster, further increasing the
planted area. In Nigeria, where peanuts are grown in
mid-slope regions on soils highly susceptible to
erosion, ox-drawn ridgers are used to control the
erosion. 117
Modern Commercial Agriculture
As population grows agricultural productivities
must be raised. Modern inputs are needed to achieve
this. Commercial fuel inputs to agriculture include
mechanized land preparation, mechanized irriga-
tion, and synthetic fertilizers. 118
The degree of mechanization varies widely, but
generally increases with per-capita income. Mecha-
nization reduces the time and labor needed for
preparing, planting, and harvesting crops. In favora-
ble areas, it also aids double cropping. The tractors
themselves come in many forms— in China the most
popular is probably the "Worker-Peasant," a 7-hp
garden tractor. In India, where the number of tractors
almost doubled from 1972 to 1977, 119 the most
popular is a 30-hp diesel.
Irrigation is most commonly done with either
electric motor or diesel driven pumps. Electric
pumps are quite reliable (although subject to inter-
ruptions in the electric power grid) and convenient,
and are often the lowest cost alternative. Diesel-
electric pumping systems, in which diesel generators
produce electricity that is then used to drive electric
pumps, and direct diesel and gasoline-powered
pumps are more often used where no electric grid is
available. These are much less mechanically de-
pendable than electric pumps.
In China, irrigation is a significant consumer of
electricity. It is estimated that 70 percent of the
electricity consumed in rural areas is for irrigation,
with the remainder used for food processing, various
rural industries, and lighting. 120 In India, the number
of electric pump sets for irrigation has grown rapidly
(see figure 3-18), and the electricity consumption for
these pump sets has gone from 4,470 GWh in
Figure 3-18-Use of Agricultural Pumpsets in India,
1950-1990
Number of pumps by type (millions)
1970 1980
+ Diesel pump sets
1990
1950 1960
"Electric pump sets
+ Animal-powered pump
SOURCE: Tata Energy Research Institute, TERI Energy Data Directory
and Yearbook (TE/2DYJI 7988 (New Delhi, India: 1989), footnote
56.
1970-71 to 23,420 GWh in 1985-86. The number of
diesel pump sets has also grown, but they still are
fewer in number than the electric units. 121
Transportation
The transportation sector accounts for a quarter or
more of total commercial energy use in most
developing countries-India and China being the
most notable exceptions (see table 3-1). Most of this
transport energy is from oil. Energy use for transpor-
tation in the developing world is expected to grow
rapidly in the future, as increasing urbanization and
incomes (see figure 3-19) lead to increased demand
for transportation services. This will increase the
outflow of scarce foreign exchange for the oi I-
importing countries, and will also require considera-
ble investment in roads and related infrastructure.
Transportation can be provided by air, rail, road,
or water. Inmost of the developing world, as well as
in the industrialized world, road technologies pro-
vide most transport services. Notable exceptions are
India and China, which have large rail networks.
In rural and poor areas of the developing world,
walking is the principal transport "technology."
The advantages of walking are many-it requires no
capital investment, it is not restricted to roads, and
117 Ibid.
118 This excludes energy used for fertilizer production and that used for crop preparation both of which we include under industrial energy use.
119 Tata Energy Research Institute, TEDDY 1988, op. cit., footnote 56, p. 137.
120C, Howe, China's Economy, op. cit., footnote 1 10, p. 88.
12l In 1983/84, for example, there were aoout 5.3 million electric pump sets and 3.1 million diesel pump sets in India. Some Of these diesel units were
used as backups for the electric units. Tata Energy Research Institute, TEDDY 1988, op. cit, footnote 56, p. 135.
Chapter 3— Energy Services in Developing Countries • 77
Figure 3-19-Passenger and Freight Transport v. GNP for Selected Countries, 1960-81
Passenger traffic and GNP per capita
10,000
o —
iE "5.
2 o
¥T5
1000
10,000
GNP per capita (1979-1981 constant use)
Total freight and GNP for selected countries, 1980-1981
1,000
1,000
GNP (BIL.US$)
This figure shows how passenger and freight transport energy use have increased with GNP for seven countries. The individual data points
are for specific years.
SOURCE: J. Venny and L. Uy, "Transport in China," World Bank staff working paper, No. 723, Washington, DC, 1985.
78 • Energy in Developing Countries
it requires no fossil fuels. On the other hand, it is
slow, tiring, and requires energy in the form of food.
Animal technologies, such as bullock carts, are
sometimes faster, have a much greater freight
capacity, and involve less work for people. Capital
investment in the bullock and cart is required,
however, as well as an operational cost for feed.
These two technologies-walking and domesticated
animals-are the principal means of transport in
many poorer and rural areas, particularly in Africa
and Asia.
Bicycles are a popular transport technology,
especially in China, where from 50 to 90 percent of
urban vehicle trips are made by this mode. 122 The
frost cost of a bicycle can be a barrier-a new bike
costs the equivalent of 7 to 8 months' wages in
Tanzania, for example-but in some areas bicycles
can be bought on credit. In India, government
employees are entitled to loans for vehicle purchase,
which can be used to buy a bicycle. Bicycles work
well in congested urban areas, where they have some
advantages over private automobiles-they are eas-
ier to park and store, less expensive to own and
operate, and do not contribute to air pollution. Their
range and freight capacity, however, are limited.
The technological leap to the internal combustion
engine allows for much higher speeds, longer
distances, larger freight capacity, and greater com-
fort. The disadvantages of the internal combustion
engine are technological complexity, movement
largely constrained to roads, high first cost and
operating cost, and environmental damage due to
fossil fuel burning. There are also secondary effects,
such as injury and death due to accidents and land
use for roads and parking. Despite its disadvantages,
the internal combustion engine is the dominant
transport technology in the industrialized and devel-
oping world, and its use is growing rapidly.
Passenger Road Transport Technologies
Mechanized passenger road transport in the devel-
oping world is performed by a wide range of
technologies, including mopeds, private autos, and
buses. Developing countries have only about 1
percent as many autos per personas does the United
States (see table 3-17), but their automobile fleets
are growing rapidly. Further, the scrappage rate (the
fraction of vehicles retired each year) is very low in
developing countries, due to the high value placed
Table 3-17 — Passenger Fleet Size and Growth
in Selected Countries
Average annual
Automobiles per growth in automobile
Country 1,000 people, 1986 fleet size, 1982-86
China 0.7 41.6
India 1.8 8.2
Kenya 8.9 3.2
Thailand 21.9 8.8
Brazil 87.0 8.9
Japan 234.0 3.0
West Germany 444.8 3.3
United States 673.4 2A
SOURCES: Fleet size and growth from Energy and Environmental Analy-
sis, Inc., "Policy Options for Improving Transportation Energy
Efficiency in Developing Countries," contractor report pre-
pared for the Office of Technology Assessment, March 1990.
Population from World Bank, World Development Report 1989
(New York, NY: Oxford University Press, 1989). Figures forthe
United States include both autos and light trucks; data are from
Oak Ridge National Laboratory, Transportation Energy Data
Book, Edition 10, ORNL-6565 (Oak Ridge, TN: Oak Ridge
National Laboratory, September 1989).
on any vehicle that runs. Therefore, choices made
now as to the energy efficiency of new vehicles in
developing countries are doubly important-these
vehicles will soon be the majority of the fleet and
they will be on the road a long time.
Although their technical efficiency (vehicle kilo-
meters traveled per liter of fuel consumed) is lower,
vehicles in developing countries average a much
higher load factor (persons per vehicle) than those in
industrialized countries. Buses are chronically over-
loaded, and mopeds and motorcycles designed for
one often carry two or more. Shared ride technolo-
gies, such as jitneys, are commonly filled beyond
rated capacity. This increases the efficiency of the
transportation system i-n terms of passenger-
kilometers per liter of fuel consumed, but reduces
safety and comfort.
The vehicles themselves are often less energy-
efficient than those found in the industrialized
world. They are often based on designs that empha-
size sturdiness and dependability under adverse
conditions (poor roads, chronic overloading, little
maintenance) over energy efficiency.
Freight Road Transport Technologies
Road freight movement in the developing world
is provided mostly by diesel trucks (with the
exception of China where much of the truck fleet
uses gasoline) and these trucks account for over half
122 World Bank, "Gridlock Weary, Some Tom to Pedal power," The Urban Edge, vol. 14, No. 2, March 1990.
Chapter 3— Energy Services in Developing Countries • 79
Table 3-18-Energy Efficiency of Trucks in Selected Countries
Energy consumption
Country/
Capacity
(megajoules per metric ton
region
Truck name
(metric tons)
per
kilometer)
OECD
Mercedes Benz 1217 (1979)
7.0
1.0
OECD
Man-VW 9136 (1980)
5.9
1.0
India
TATA 1201 SE/42
5.0
2.1
India
Ashok Leyland Beaver
7.5
1.6
China
Jiefang CA-10B
4.0
2.3
China
Dongfeng EQ140
5.0
1.8
NOTE: OECD and Indian trucks use diesel, Chinese trucks use gasoline.
SOURCE: J. Yenny and L. Uy, World Bank, "Transport in China," staff working paper No. 723, 19S5, p. 70.
the energy used for road transport in the developing
world. 123 The movement of height is required for
most economic activity, and in many developing
countries the prices of diesel fuel are kept lower than
gasoline prices. In the United States, for example,
gasoline and diesel prices at the pump are almost the
same, while in India diesel is slightly less than half
the price of gasoline. 124
Trucks in the developing world are relatively
inefficient, requiring 1.5 to 2.5 times as much energy
to move one ton of freight one kilometer as
comparable trucks in the OECD countries (see table
3-18). In developing countries, however, trucks
must cope with more difficult operating conditions:
the roads are typically congested and poorly main-
tained, aggravating technical inefficiency and accel-
erating wear.
Rail Technologies
Railroads are significant providers of transport
services only in India and China, and in these two
countries the rail share of total transport is declining
rapidly due to the much faster growth of road
transport. In China, for example, the share of
passenger traffic using railways dropped from 69
percent in 1965 to 48 percent in 1987. Railway
freight transport shows the same trend of decreasing
relative use. 125 Similarly, India shows a mode shift
toward roads and away from rail for both passenger
and freight traffic. 126
Despite these modal shifts, the rail systems in
both countries still account for significant energy
use. China and India have extensive rail networks
that consume, respectively, 72 percent and 29
percent of transportation energy (see app. 3-A). The
Indian rail system, although in relative decline, still
carries a significant amount of freight and passen-
gers (see table 3-19), using a mix of steam (being
phased out), diesel, and electric locomotives.
Implications for Energy Demand
Road transport-private autos for passengers and
trucks for freight-has become the dominant mode
of transportation in developing countries. Increases
in population, income, and auto ownership rates
(autos per person) combine to yield a rapid increase
in the number of private vehicles. Increasing urbani-
zation leads to greater congestion, which reduces the
efficiency of private vehicles. Urbanization, eco-
nomic growth, and industrialization require large
increases in freight movement, as producers move
farther from markets. The net effect of these factors
will be an increase in the energy needed to provide
transportation services.
Improvements in the energy efficiency of devel-
oping world transport systems can be made in
several areas. Road-going vehicles in the developing
world are less energy-efficient than comparable
vehicles in the industrialized world, suggesting that
efficiency gains can be made in the vehicles
I23 Trucks account for 50 to 75 percent of energy consumed f or roaa transport in the developing world, compared to 30 to 35 percent for many
industrialized countries. Clell G. Harrsl, "Meeting the Transportation Aspirations of Developing Countries: Energy and Environmental Effects,"
Proceedings of the Energy and Environment in the 21st Century Conference (Cambri'dge, MA: Massachusetts Institute of Technology, March 1990).
124 Energy Information Administration, International Energy Annual 1988, DOE/EIA-0219(88) (Washington, DC: U.S. Government Printing Office,
1989).
I25p Kuirun and S. Guojie, < < Overview of Transport Development in China, ' ' paper presented at the New Energy Technologies Transportation and
Development Workshop, Ottawa, Canada, September 1989.
126 JoyDunkerley, Irving Hoch, Cnaru Gadhok, Kapil Thukral, "Energy and Transport — The Indian Experience,' Pacific and Asian Journal ofEnergy,
1987, pp. 1-12.
80 • Energy in Developing Countries
Table 3-19-Comparison of Rail Systems in China,
India, and the United States
United
China India States
Length of rail network (km) 53,000 62,000 235,000
Rail energy use (percent of
total transport energy use) .... 51 27 3
Percent of freight traffic carried
by rail 45' 47 30
Percent of passenger traffic
carried by H-l 55 22 l_
SOURCES: P. Kuirun and S. Guojie, "Overview of Transport Development
in China," paper presented at the New Energy Technologies
Transportation and Development Workshop, Ottawa, Canada,
September 1989; Tata Energy Research Institute, TERI
Energy Data Directory and Yearbook (TEDDY) 1988 (New
Delhi, India: 1989); International Energy Agency, Worid En-
ergy Statistics and Balances; 198 f-87 (Paris: OECD, 1989);
Oak Ridge National Laboratory, Transportation Energy Data
Book: £S/on JO, ORNL-6565, 1989; Association of American
Railroads, Railroad Facts 7989 (Washington, DC: American
Association of Railroads, 1990).
themselves. Improvements in the transportation
infrastructure, such as improved roads and reduced
congestion, can also increase energy efficiency.
Mode choices, such as a movement away from
private autos and motorcycles to buses and bicycles,
can help. Of course, all these options have benefits
and costs: these will be explored later in this OTA
study. The important conclusions for this section,
however, are that the demand for transportation
services is increasing rapidly, technologies in use
today are not as efficient as they could be, and the
energy impacts of technology choices made today
will be felt far into the future.
Conclusion
This survey of energy services and how they are
provided in developing countries reveals three
common characteristics. First, each service is pro-
vided by a wide range of technologies and fuels.
Cooking is provided by technologies ranging from
open frees to microwave ovens, with a large number
of possibilities in between. The range of passenger
transport services is similarly wide, varying from
foot to jet passenger airplanes. Second, there is in
almost all cases a well-established transition be-
tween technologies, depending on two main factors-
income and availability of fuel supplies. Third, the
services are currently being provided by technolo-
gies whose efficiency could usually be significantly
improved.
The following chapter will examine how the
many types of energy used in developing countries-
fossil fuels, electricity, and biomass fuels-are
provided, including domestic production, imported
supplies, the energy distribution system, and the
energy conversion sector.
Chapter 3~Energy Services in Developing Countries .81
Appendix 3-A — Energy Balances for
Selected Developing Countries
The following energy balances begin with Interna-
tional Energy Agency (IE A) 127 energy production, trade,
and stock change totals for commercial fuels. This
provides a common framework for evaluating individual
countries and for comparing different countries. Biomass
fuels for the traditional sector are not included in the
country-specific IEA data and so are separately added
based on country specific field survey data. The year
chosen for each country is determined primarily by the
year for which the biomass energy data is available.
In contrast to the IEA procedure, energy supply
production, conversion, and transformation losses are not
separately tallied in the energy balances presented here.
Instead, these losses are carried forward into the sectoral
breakdowns in proportion to the IEA sectoral breakdown
of energy use. This more accurately indicates sectoral
energy usage by showing the losses incurred in providing
energy to each sector.
Electricity is initially divided into two categories in the
following energy balances: nonthermal and thermal.
Nonthermal electricity is given in terms of the electric
power output--the joule equivalent of kWh. Thermal
electricity is given in terms of thermal energy input; losses
incurred in generation, transmission, and distribution are
kept in the total. Nonthermal and thermal electricity
quantities, therefore, can not be directly compared.
The IEA convention for electricity production divides
the hydroelectric output in kWh by 0.385 in order to make
hydroelectric power appear to be on the same 'thermal
equivalent" input basis as thermally generated electricity
when listed on the basis of fossil fuel input. The IEA
subsequently multiplies the sum of hydroelectric "ther-
mal equivalent" and thermal electric inputs by 0.385 to
get an electric power output in kWh. Thermal and
hydroelectric "thermal equivalent" losses are lumped
together as an energy production loss.
This convention of "thermal equivalents" leads to a
large misrepresentation in the energy balances for hydro-
rich countries such as Brazil. The procedure used here
avoids the IEA convention of assigning a thermal
equivalent for hydroelectric or other nonthermal power. It
also carries the losses in thermal generating plants
through to the end-use sectors as noted above. At the
sectoral level, the thermal and nonthermal electricity are
added together directly to indicate the average amount of
energy, including fossil fuel, used by each sector. These
figures are shown in brackets to denote that the figure is
a sum of nonthermal electricity output and thermal energy
input. This procedure lowers the energy supply totals
compared to those usually found in the literature.
Percentage breakdowns by end-use sector are based on
the IEA data; percentage breakdowns by energy service
within end-use sectors are based on country-specific
surveys as noted. The end service breakdowns are the best
estimates that OTA could make given the poor quality and
paucity of available data These breakdowns are provided
here only as an indication of the relative importance of
selected energy services; they should not be construed to
be a precise quantitative measure of the energy consumed
in delivering these services or to be a precise listing of
energy services and their interrelationships. Some impor-
tant energy services and fuel mixes are overlooked in
many of the available energy service breakdowns. For
example, lighting and the use of traditional fuels are
largely left out of the industrial sector. In addition, a
number of important energy services are generally left out
of the breakdowns: an example might be the use of
animals for traction in agriculture and the use of crop
residues to feed them.
'^International Energy Agency, World Energy Statistics and Balances: 1971-87 (Paris: OECD, 1989).
82 • Energy in Developing Countries
Table 3A-I — Brazil: Energy Supplies and Services, 1987 Exajoules (10 18 Joules= 0.9478 Quad)
and Percent of National Total
Fossil fuels Eleotricity Blomass
Cane/ Wood/
Energy supplies COal Oil Gas Nonthermal Thermal alcohol' Charcoal Other Total
Production 0.124 1.55 0.112 0.681 — 0.91 1.39 0.13 4.90
Trade/stock change 0.315 1.040 — 0.023 — — — -0.01 1.37
Electric generation -0.044 -0.073 — — 0.117 — — — 0.00
Nonenergy — -0.167 — — -0.02 — — -0.19
Total energy 0.40 2.349 0.112 (0.821) 0.89 1.39 0.12 6.08
Percent of total 6.6% 38.6% 1.8% 14.6% 22.9% 2.0% 100'YO
Energy services
Residential — 0.202 0.007 (0.169) — 48 — 0.86
Residential — 3.3% 0.1% 2.8% — 7.9% — 14.1%
Cooking/heating 3.3% 0.1% 0.77%* — 7.9% — 1 2.1%
Lighting — — — 0.67% _ _ _ n 7<>/o
Refrigeration — — — 0.90% _ _ _ n g%
Television — — 0.1 7% — — — 0.2%
Air conditioning — — 0.09% — — — 0.170
Other — — — 0.21 % _ _ _ n 2 o /o
Commercial' — 0.026 0.003 (0.159) — 0.018 — 0.21
Commercial — 0.42% 0.05% 2.62% — 0.3% — 3.5%
Cooking/heating 0.42% 0.05% 0.20% — 0.3% — 1 .0%
Lighting — — — 1.1570 _ _ _ 1 270
Refrigeration — — — 0.45% — — 0.5%
Air conditioning — — 0.52% — — 0.5%
Other — — 0.29% _ _ _ n 3%
Industrial 40 0.56 0.10 (0.46) 0.44 0.75 — 2.71
Industrial 6.58% 9.22% 1.65% 7.57% 7.24% 12.3% — 44.6%
Motor drive — — 3.7170 — — 3.7%
Process heat 6.5% 9.22% 1.65% 3.18% 7.24% 12.3% — 40.2%
Lighting — — — o.15?40 _ _ _ 2 o /o
Electrochemical — — 0.53% _ _ _ n 5<>/o
Transport 000, 1.417 — (0.005) 0.44 — ■ 1.86
Transport |l) 23.3% — 0.08% 7.24% — — 30.6%
Road — 20.4% — — 7.24% — — 27.6%
Rail ° * 0.43% — 0.08% _ _ _ n 5 <, /o
Air — 1 .74% — — — — — 1 .7%
Other — 0.77% — — _ _ _ n Q%
Agriculture — 0.146 — 0.026 — 0.126 — 0.30
Agriculture — 2.4% — 0.43% — 2.1 % — 4.9%
—Not available or not applicable.
() data in parentheses is sum of nonthermal energy output and thermal energy input.
a The use of bagasse for energy production(cogeneration)isdivided proportionately betweenindustrialp process heat and road transport. Electricity generation
within the cane industry is not given separately.
b This is mostly for water heating(10 TWh). Only 0.5 TWh were for cooking.
'Excludes public buildings.
SOURCES: Adapted from international Energy Agency.lVorid Energy Statistics and Balances: 1971 -87 (Paris: OECD, 1989); and Brazilian Ministry of Mines
and Energy, National Energy Balance forBrazil1988 (Brasilia, 1988), provided by Howard S. Geller, American Council for an Energy Efficient
Economy, Washington, DC, and Sao Paulo, personal communication, Mar. 8,1990.
Chapter 3™Energy Services in Developing Countries • 83
Table 3A-2-China: Energy Supplies and Services, 1987 Exajoules (10" Joules = 0.9478 Quad)
and Percent of National Total
Fossil fuels Electricity Blomass
Energy supplies coal Oil Gas Nonthermal Thermal Crop Wood
Production f8.l'5 5.32 6.46 6.33 — 3.4 3.2
Trade/stock change . . -0.80 -1.40 — — 0.00 — —
Electric generation .. . -3.15 -0.59 -0.02 — 3.76 —
Nonenergy -0.15 — —
Total energy 14.20 3.18 0.44 (4.10) 3.4 3.2
percent of total 49.5% 11.170 1.570 14.3% 11.870 11.2%
Energy services
Residential 5.23 0.12 0.07 (0.27) 3.3 3.2
Residential 18.2% 0.4% 0.2% 0.94% 11.5% 11.2%
Cooking 8.7% — 0.2% — 11.5% 10.5%
Space heating 9.5% — — — 0.7%
Lighting — 0.4% 0.94%
Commercial/public . . . 0.31 0.27 0.00 (0.19) —
Commercial/public... 1.1% 0.9% — 0.66% — —
Industrial 9.41 1.77 0.37 (3.06) 0.1 —
Industrial 32.8% 6.2% 1 .3% 10.7'YO 0.3% —
Process heat 32.8% 3.4% 1 .3% — 0.3% —
Mechanical drive . . — 2.8% — 10.7% — —
Transport 0.59 0.57 0.00 (0.08) — —
Transport 2.1 % 2.0% - 0.285 —
Road — 0.8% — — —
Rail 2.1 % — — 0.28% — —
Air — — — — — —
Other 0.00 1 .2% 0.00 — —
Agriculture 0.98 0.44 — (0.50) — —
Agriculture 3.4% 1.570 — 1.7% — —
— Not available or not applicable.
( ) data in parentheses is sum of nonthermal energy output and thermal energy input.
SOURCE: Adapted from International Energy Agency, World Energy Statistics and Balances: 1 971 -87 (Paris: OECD,
Energy," contractor report prepared for the Office of Technology Assessment, 1990.
fuels
Dung Biogas Total
0.15
0.03
31.0
-2.2
0.15
0.5%
0.03
0.1%
-0.15
28.7
100%
0.15
0.5%
0.4%
0.1%
0.03
0.1%
0.1%
12.4
43.%
31 .4%
10.3%
1.370
0.77
2.7%
—
—
14.7
51 .3%
37.8%
13.5%
—
—
1.24
4.3%
0.8%
2.4%
—
—
1.270
—
1.92
6.7%
1989); and Vadav Smil, "China's
84 •Energy in Developing Countries
Table 3A-3-lndia: Energy Supplies and Services, 1985 Exajoules (10" Joules = 0.9478 Quad)
and Percent of National Total
Fossil fuels Electricity Biomass
Energy supplies coal Oil Gas Nonthermal Thermal Wood Dung Crop Total
Production 3.16 1.29 0.17 0.20 — 0.87 1.2 1.6 8.5
Trade/stock change 0.03 0.48 — — 0.00 _ _ _ q g
Electric generation -1.29 -0.11 -0.05 — 1.45 _ _ _
Nonenergy — -0.09 — — — — — -0.09
Total energy 1.90 1.57 0.12 (1.65)/19% 0.87 1.2 1.6 8.9
percent of total 21% 18% 1.3% 10% 13% 18% 100%
Energy services
Residential 0.06 0.39 0.007 (0.20) 0.78 1.2 1.6 4.2
Residential o.7% 4.4% 0.0870 2.2% 8.8% 14% 18% 47%
Cooking/water heating 0.6% 1.9% 0.08% — 8.8% 14% 18% 4370
Lighting 2.1% — 1.8% — ■ — 3.9%
Appliances — — — o.4% — — 0.4%
Commercial/pubiic 0.02 0.03 (0.09) 0.05 — '„ 0.20
Commercial/pubiic 0.2% 0.3% 0.6% — , 2.2%
Cooking/heating 0.2?! 0.3% — m 0.6% — 1.1%
Lighting — — — 0.4% — — 0.4%
Appliances — — 0.6% — — 0.6%
industrial 1.61 0.40 0.113 (0.97) 0.04 — — 3.1
industrial 18% 4.5% 1.3% 10.9% 0.5% — — 35%
Process Heat 18% 3.1% 1.3% — 0.5% — — 23%
Motor Drive — 0.8% — 10.0% — — — 10.8%
Lighting — — 0.5% — — 0.5%
Appliances — — 0.5% — — 0.5%
Transport 0.23 0.77 — (0.04) — — — 1.0
Transport 2.6% 8.7% 0.5% — — — 11.8940
Road 7.1% — — — — 7.1%
Rail 2.6% 0.7% — 0.5% — — 3.8%
Air : 0.8% — — — — — 0.8%
Agriculture" — 0.15 0.003 (0.28) — — — 0.43
Agriculture — 1.7% 0.03% 3.1% — — — 4.8%
Motor Drive — 1.0% — 3.1% — — 4.1%
Traction — 0.7% — — — — — o.7%
—Not available or not a aplicable.
() data in parentheses is sum of nonthermal energy output and thermal energy input.
• Small.
a baseline data from the international Energy Agency for petroleum use in agriculture have been modified to correspond better with TERI energy data.
SOURCE: Adapted from international Energy Agency, World Energy Statistics and Balances: 1971 -87 (Paris: OECD, 1989); Tata Energy Research
Institute, TEDDY, op. cit., footnote 58; and AshokDesai, contractor report prepared for the Office of Technology Assessment and personal
communication.
Chapter 3-Energy Services in Developing Countries »85
Table 3A-4-Kenya: Energy Supplies and Services, 1980, Petajoules (10 5 Joules) and Percent of National Total
Fossil fuels Electricity Blomass fuels
Energy supplies coal Oil Gas Nonthermal Thermal Wood' Residues Total
Production — — ~ 3.82 — 320 9.3 333
Trade/stock change 0.42 78 — 0.44 — — 79
Electric generation — -6.1 — 6.1 — 0.0
Nonenergy -2.3 — — — — — -2.3
Total energy 0.42 70 — (10.4) 320 9.3 410
percent of total 0.1% 17.1% — 2.5% 78% 2.3% 100%
Energy services
Residential — 6.84 — (2.19) 253 9.3 271
Residential — 1.67% — 0.53% 62% 2.3% 67%
Cooking — 0.53% — 0.32%. 62%. 2.3%. 65%
Lighting — 1.14% — 0.21% — — 2%
Commercial — 0.75 — (1.92) 3.5 — 6.2
Commercial — 0.18% — 0.47% 0.85% — 1.5%
Cooking/heating — 0.05% — — 0.85% — 0.9%
Lighting — 0.13% _ 0.47% — — 0.6%
Industrial 0.42 16.5 — (3.23) 56 — 76
industrial 0.1% 4.0% — 0.79% 13.6 — 18.5%
Informal — — 9.4% — 9.4%
Formal 0.1% 4.0% — 0.79% 4.2% — 9.1%
Transport — 43.3 — — 43.1
Transport — 10.6% — — 10.6%
Road — 6.9% — — 6.9%
Rail — 0.6% — — — — 0.6%
Air ■ 2.8% — — — — 2.8%
Agriculture — 6.6 — (1.06) — 7.7
Agriculture — 1.6% — 0.26% — 1.%
() data in parentheses is sum of nonthermal energy output and thermal energy input.
a Includes both commmercial and noncommercial „„„,„.,„„,.. include wooer used as a feedstock or as a construction material. .Alsoincludes charcoal
that is produced from wood. This conversion takes roughly 110PJ of wood and converts it into about 27 PJ of charcoal, of which about 1.3 PJ is lost during
distribution.
SOURCE: Adapted from International Energy Agency, World Energy Statistics and Balances: 1971 -87 (Paris: OECD, 1989); and "Energy and Development
in Kenya," Eds. Phil O'Keefe, Paul Raskin, and Steve Bernow,Beijer Institute, Royal Swedish Academy of sciences, Stockholm, Sweden, 1984.
86 • Energy in Developing Countries
Table 3A-5-Taiwan: Energy Supplies and Services, 1987, Petajoules (101 'Joules) and Percent of National Total
Fossil fuels Electricity Blomass
Energy supplies coal Oil Gas Nonthermal Thermal wood" Total
Production 38.9 5.4 42.3 145 — 21 253
Trade/stock change 363 758 — — — 1121
Electric generation -220 -58.3 — — 278
Nonenergy -46.8 — — ■ -47
Total energy 182 658 42.3 (423) 21 1326
percent of total 13.7% 49.6% 3.2% 31.9% 1.6% 100.%
Energy services
Residential 0.14 46.3 19.6 (87.5)/6.6% 21 175
Residential |Y0 3.5% 1.5% 1.6% 13.2%
Cooking/water heating ... . — 3.5% 1.5% 0.3% 1.6% 6.9%
Lighting — — — I.O% — 1.0%
Refrigeration — — — 1 .9% 1 .9%
Television — — — 1 .2% — 1 .2%
Fans/air conditioning — — — 1.5% — 1.5%
Commercial/public 0.12 31.6 4.0 (&.22 0.3 83.2
Commercial/public • 10 2.4% 0.3% — 6.3%
Cooking/water heating .... 11/0 2.4% 0.3% 0.1% •% 2.8%
Lighting — — — 1.2% 1.2%
Air conditioning — — — 1.1% 1.1%
Other Appliances — — — 1.2% — 1.2%
Industrial 180 314 18.5 (255) — 767.5
Industrial 13.6% 23.7% 1.4% 19.2% — 57.9%
Transport — 222 — (2.7) 225
Transport — 16.7% — 0.2% 17.0%
Road — 15.0% _ _ _ 15.0%
Rail — 0.1% — 0.1% — 0.2%
Air — 1.0% — * — 1.0%
Other — 0.6% — ' 0.6%
Agriculture — 37.3 — (13.0)/1.0% — 50.3
Agriculture — 2.8% — — 3.8%
—Not available or not applicable.
() data in parentheses is sum of nonthermal energy output and thermal energy input.
• Small.
aCharcoal is includedunder wood. The charcoal conversion effiency is assumed to be a relatively high level of50 percent by energy.
SOURCE: Adapted From International Energy Agency, World Energy Statistics and Balances: 1971 -37 (Paris: OECD, 19S9); and personal communication,
Dr. Gwo-Tzeng, Energy Research Group and Institute of Traffic and Transportation, NationaChiao Tung University, Taipei, Taiwan.
Chapter 3— Energy Services in Developing Countries • 87
Table 3A-6-United States: Energy Supplies and Services, 1985,'Exajoules (10 s Joules =0.9478 Quad)
and Percent of National Total
Fossil fuels Electricity Biomass*
Energy supplies coal Oil Gas Nonthermal Thermal Wood 1 Total
Production 19.6 20.9 16.5 2.53 ' — 2.80 62.3
Trade/stock change -1.8 8.9 1.2 006 " — — 8.4
Electric generation -14.8 -1.1 -3.0 — 18.9 -0.02 —
Nonenergy — -1.8 — — — -1.8
Total energy 3.0 26.9 14.7 (21.5) 2.8 68.9
percent of total 4.4% 39% 21.3% 31.2% 4.1% 100%
Energy services
Residential 0.1 1.5 5.3 (7.6) 1.0 15.5
Residential 0.1% 2.2% 7.7% 11.% 1.5% 22.5%
Space conditioning — 2.0% 5.3% 4.3% 1.5% 13.1%
Water heating — 0.2% 1.8% 1.5% — 3.5%
Cooking — — 0.5% 0.7% — 1 .2%
Refrig/Freezers — — — 1 .6% — 1 .6%
Lighting/Other' — — 0.1% 2.8% — 2.9%
Commercial/public 0.1 1.2 2.9 (6.6) — 10.8
Commercial/public 0.2% 1.7% 4.2% 9.6% — 15.7%
Space conditioning — 1.6% 2.9% 6.1% — 10.6%
Water heating — 0.1% 0.1% 0.1% — 0.3%
Lighting — — — 2.5% — 2.5%
Industrial 2.6 4.3 6.5 (7.3) 1.8 22.5
Industrial 3.8% 6.2% 9.4% 10.6% 2.6% 32.7%
Heat' 3.8% (11.8%)9 — 1.2% 2.6% 19.4%
Motor Drive — — — 7.1% — 7.1%
Off-Highway transport .... 0.9% — — — — 0.9%
Transport — 19.3 — (0.03)/'% — 19.3
Transport — 28.% — — 28%
Road — 23.1% — — — 23.1%
Rail — 0.7% — •% — 0.7%
Air — 3.9%0 — — — 3.9%
Other — 0.3% — — — 0.3%
Agriculture" — 0.6 — — — 0.6
Agriculture — 0.9% — — — 0.9%
—Not available or not applicable.
()data In parentheses is sum of nonthermal energy output and thermal energy input.
"Small.
aNoto that the IEA data used as a framework for this energy balance differs slightly from official U.S. energy statistics. See, for example, Energy Information
Administration, Month/yGwrgy Review, January 1990.
b split 58 percent Nuclear and 42 percent Hydroelectric and other (geothermal, solar, etc.).
d Imports of electricity into the United States are primarily hydroelectric based power from Canada,
e Includes clothes washers and dryers, dishwashers, lighting, and miscellaneous,
f Includes fuels US e d for cogeneration applications.
gOil and gas applications are combined here,
h This does not include indirect inputs (fertilizer, etc.).
SOURCE: Adapted from International Energy Agency, WortdEnergy Statistics and Balances: 1971 -87( Paris: OECD, 19S9); and Paui D. Holtberg, Thomas
J. Woods, Marie L. Lihn and Nancy C. McCabe, Baseline Projection Data Book: 1989GRI Baseline Projection of U.S. Energy Supply and Demand
to 2010 (Washington, DC: Gas Research institute, 1990).
88 • Energy in Developing Countries
Appendix 3-B — Sources for Table 3-9
As used here, efficiency is approximately the first-law
efficiency; that is, the total heat transfer to the material
being processed divided by the heat input to the system.
This is to be contrasted with the second-law efficiency,
which compares the heat transfer achieved in the process
with the maximum possible heat that could be transferred
for the same purpose by any system using the same energy
input. For more information, see Efficient Use of Energy,
MP Conference Proceedings, No. 25, American Institute
of Physics, New York, 1975.
Cooking: The primary source for traditional and
modern cooking technologies is Samuel F. Baldwin,
Biomass Stoves: Engineering Design, Development, and
Dissemination, op. cit, footnote 17. An enormous range
of efficiencies have been reported for the open fire,
ranging from 2 to 30 percent. A list of some 45 reports on
traditional cooking technology efficiencies was devel-
oped byj as Gill in 1981 and is cited in K. Krishna Prasad,
Woodburning Stoves: Their Technology, Economics, and
Deployment (Geneva: 1983). International Labor Organi-
zation, World Employment Programme Research. Most
of these reports, however, do not cite a source, nor any
details as to how such an efficiency figure was derived.
Traditional Beer Brewing: Data are from Frans
Sulilatu, "Improved Beer Cookers In Burkina Faso,"
Boiling Point, No. 10, pp. 10-14, August 1986. This is the
thermal efficiency of heating the brew to boiling, not for
the entire brewing process. In Burkina Faso, West Africa,
traditional dolo (beer) cookers using 80-liter clay jars
have thermal efficiencies of 17 percent with a power
output of 80 kW for a cooker with four jars arranged in a
square, and 15 percent at 35 kW with the jars arranged in
a line with fires between adjacent jars. Simple improve-
ments in stove design and the use of aluminum pots raised
efficiencies up to as high as 53 percent.
The efficiency of the brewing process can also 6
determined by calculating the total amount of useful
energy provided. Mnzava, cited below, has estimated that
0.12 to 0.5 nfof fuelwood are needed to brew 100 liters.
Assuming that 1 nfof stacked fuelwood weighs 500 kg
and has an energy content of 16 M.J /kg for a total of 8 GJ ;
and assuming that the energy is used to bring the water to
a boi I once to steri I ize it and then to mai ntai n its
temperature at a low level while it ferments; the energy
required to heat 100 liters of water to a boil from ambient
(20 'O is 100(80)4.186=14 MJ.This gives an efficiency
range of 0.3 to 0.7 percent for the beer-brewing process.
This very low efficiency compared to the thermal
efficiency found for dolo cookers in Burkina Faso maybe
due to the long, low temperature heating needed to
concentrate the mash as well as for other operations. See:
EMMnzava, "Fuelwood and Charcoal In Africa," W.
Palz, P. Chartier, and D.O. Hall (eds.), Energy From
Biomass, 1st E.C. Conference on Biomass, Brighton, East
Sussex (London: Applied Science Publishers, Ltd., 1980).
Modern Beer Brewing: Data are from Bernard B.
Hamel et al., "Energy Analysis of 108 Industrial Proc-
esses," 1980. The figure of 79 percent is theoverallboiler
efficiency for a modern brewery. This boiler provides
process steam, hot water, and other heating services. The
efficiency of the cooking process is somewhat reduced
from this level, but no separate estimate was available.
The value of 6 percent is based on a total energy
demand for a brewery found by Hamel et al„ of 1,439 Btu
per pound of beer produced or, equivalent^, 3.35 MJ/kg
compared to the energy required to heat the brew to
boiling a single time to sterilize it— as in the comparison
made for the traditional technology.
Tobacco Drying: The estimate of tobacco dryi ng
efficiency is from M.J. Mwandosya and m.l. Luhanga,
Energy Use Patterns In Tanzania, Center for Energy and
Environmental Studies, Princeton University, Report No.
180, February 1985; and M.J. Mwandosya and MX.
Luhanga, "Energy Demand Structures in Rural Tanza-
nia,' Department of Electrical Engineering, University of
Dar-es-Salaam, Tanzania, 1984.
Traditional Tea Drying: Tea drying is based on the
data in Mwandosya and Luhanga, listed above. They
estimate that 150 kg. of green tea requires 9.4 GJ of
fuelwood, resulting in 30 kg. of dried tea. To evaporate
120 kg of water requires 2,260( l20)kJ =271 MJ of
energy, for an efficiency of 2.9 percent.
Traditional Baking: For traditional bakeries, Ahmed
and Elamgzoub found 0.5 to 0.8 kg of wood used per kg
of flour. Typical ratios for bread are 720 g flour, 500 g
liquid, and 50 g sugar input per kg of bread output. With
specific heats of 1.8 kJ/kg 'C for flour, 4.186 U/kg 'C for
water, and arbitrarily assuming 4.0 kJ/kg C for sugar; and
noting that approximately half the water evaporates, the
rest remaining in the bread (Geller); then the energy
required to bake 1 kg of flour into bread at 190 'C is:
(1.0)(1.8)(170) + (0.7)(4.186)(80) + (0.35)(2260) +
(0.35)(4.186)(90) + (0.09)(4.0)(170)= 1.5 MJ, where it
was assumed that the specific heat of the water remaining
in the bread, 0.35 kg, remained 4.186 and the chemical
reactions and heating of the vaporized steam were
ignored. By comparison, 0.5 to 0.8 kg wood have an
energy content of 8 to 13 MJ. Abdel Salaam Ahmed and
El Sheikh Elamgzoub, Survey of Fuelwood Consumption
in Khartoum Province Industries ( K h a rtOU m, Sudan:
National Energy Administration, Ministry of Energy and
Mining for the Energy Research Council, Sudan Renew-
able Energy Project, April 1985. Howard S. Geller and
Chapter 3— Energy Services in Developing Countries • 89
Gautam S. Dutt, "Measuring Cooking Fuel Economy",
Wood Fuel Surveys (Rome: 1983). Food and Agriculture
Organization of the United Nations, GCP/INT/365/SWE.
Reddy and Reddy found that 0.583 kg wood were used
to cook 1 kg of maida, corresponding to an efficiency of
16 percent. Amulya Kumar N . Reddy and B. Sudhakar
Reddy, "Energy Use in a Stratified Society: Case Study
Of Firewood, in Bangalore,' Economic and Political
Weekly (India), vd. 18, No. 41, Oct. 8, 1983.
Shirey and Selker list the efficiencies of a number of
traditional and modem ovens used in a variety of
countries. Ovens in Somalia, Sudan, Guatemala, Zim-
babwe, and Sri Lanka have typical measured efficiencies
of 1 to 3 kg wood per kg flour, giving efficiencies, as
calculated above, of 3 to 8 percent. In contrast, an
improved wood-fired Somali oven is cited as using 0.16
kg of wood to cook 1 kg of flour into bread-an efficiency
of 58 percent; and modem natural gas ovens are listed as
baking 360 kg of flour into bread using 1 GJ of energy
-an efficiency of 54 percent. E . Shirey and J . Selker,
"Bread Ovens," Boiling Point, No. 10, pp. 18-21,1986.
Modern Bakeries: Ho, Wijeysundera, and Chou found
first-law efficiencies for a modem industrial bakery in
Singapore to be 43 percent for the entire process,
including preparation of the dough. Second-law efficien-
cies were also calculated and found to be 15.5 percent.
J.C. Ho, N.E. Wijeysundera, and S.K.Chou, "Energy
Analysis Applied to Food Processing," Energy vol 11,
No. 9, 1986, pp. 887-892.
Fish Smoking: Mwalyosi estimates that smoking 1 kg
of fresh fish requires 4 to 5 kg dry wood. If 70 percent of
the fish is assumed to be water, then it requires (2,260
kJ/kg)(0.7 kg)= 1.6 mj to evaporate the water compared
to (4 to 5 kg)(16 MJ/kg) = 64 to 80 MJ of wood to
accomplish the task, for an efficiency of 2.0 to 2.5 percent.
Raphael B. Mwalyosi, "Management of the Mtera Reser-
voir in Tanzania," AMBIO,voi. 15, No. 1, 1986, pp.
30-33.
Traditional Brick Firing: Schmitt estimates 1.36 MJ
of energy is required per kg of brick produced in order to
evaporate moisture from the raw brick (after drying in the
sun) and heat it to a firing temperature of 850 'C, and an
additional 0.2 to 0.4 MJ/kg is needed for the chemical
reactions. Based on observations at six sites, an average
of 2.5 MJ fuelwood and other organic matter were used
per kg of brick produced, for an efficiency of (0.2 to
0.4)/2.5 =8 to 16 percent. It should be noted that these
results were for very large kilns, firing typically 100,000
bricks at a time. Klaus Schmitt and Werner Siemers,
Energy From Agricultural Residues and Energy Utiliza-
tion In Small Scale Industries In The Sudan, Section 5.4,
"Brick Kilns" (Gottingen, Sweden: for the National En-
ergy Administration of Sudan, Khartoum, September
1985).
Gandhi found an efficiency of 6.4 percent for brick
kilns in India, representing the irreversible reactions that
take place during firing. The overall heat balance found by
Gandhi fora Bull' strench was: energy in= 3.88 MJ-HD.29
MJ in carbon in brick; energy out is 61.4 percent in dry
exhaust; 16.9 percent in moisture in exhaust; 6.4 percent
in irreversible reactions; 4.0 percent in heat loss of CO;
0.3 percent in carbon in ash; and other heat losses (by
difference) of 11 percent-presumably, much of this loss
was through the kiln walls. Other types of kilns require
from 2 to 18 MJ/brick for firing. With an average brick
size of 108 in3 or 108(16.387) cm3 and an average brick
density of 1,800 kg/m3, this gives an energy requirement
Of 2 to 18 MJ/3.18 kg or 0.637 to 5.7 MJ/kg. Sunita
Gandhi, 'The Brick Industry in India: Energy Use,
Tradition and Development," Ph.D. Thesis, Trinity
College, Cambridge, October 1986.
The brick and tile industry in Uganda uses 0.5 to 1.8
stacked cubic meters of wood per metric ton of brick
produced; with 7,650 MJ/m 3 for eucalyptus at 510
kg/stacked nfto give, at best, 3,800 MJ per metric ton of
brick output. Potential energy savings of 35 percent may
be possible simply with better fiig techniques and kiln
construction, and by the introduction of small cavities and
organic materials into the brick to reduce mass and
improve the uniformity of firing. Using the figures for
Sudan, this gives an efficiency of about 5 to 10 percent
when assuming the chemical reactions need 0.2 to 0.4
MJ/kg; using the figures for India this gives an efficiency
of about 2 percent. "Uganda: Energy Efficiency Improve-
ment in the Brick and Tile Industry,' World Bank/UNDP
Energy Sector Management Assistance Program, March
1989.
Modern Brick Industry: Assuming the same range as
for Sudan, that irreversible chemical reactions for the
process are 0.2 to 0.4 MJ/kg of fired brick, a modem brick
factory has an efficiency of 6-11%. The relatively high
observed efficiency of the traditional process relative to
modem kilns is largely due to substantial underfiring in
traditional kilns and corresponding low-quality product.
Calculated from Bernard B. Hamel et al. "Energy
Analysis of 108 Industrial Processes," op. cit.
Traditional Foundry Work: in Indonesia, an estimated
1 kg of charcoal is used per kg of aluminum melted and
Cast into pots. From the CRC Handbook of Chemistry and
Physks,the melting point of aluminum is 933 OK and its
specific heat varies linearly with temperature from C =0.9
kj /kg'Cat300'K to 1.19kJ/kg'Cat933'K. The energy
needed to heat it to its melting point is then given by
MC DT=658 kJ/kg. To melt the aluminum requires an
additional" 398 kJ/kg (CRC Handbook). The total process
33-718
0-90-4
90 •Energy in Developing Countries
then requires 1,056 U/kg. Charcoal has a calorific value
of about 33 M.T/kg. The process is therefore about 3
percent efficient. World Bank, Indonesia: Issues and
Options in the Energy Sector, UNDP/World Bank E nergy
Sector Assessment Program Repor& No. 3543 -I ND, No-
vember 1981.
Modern Foundry work: Figure of 40 percent is from
Bernard B. Hameletal., "Energy Analysis of 108
Industrial Processes,' op. cit. above, p. 282.
Chapter 4
Energy Supplies in the
Developing World
Contents
Page
introduction and Summary 93
The Overall Primary Fuel Mix 93
C_ Q I * * ** * ** * ** ** ** ** * * ** QC
Uen ) ) VJ ) +■ J Jo» ■ 5 58 ) 4 t o" J o a 30
/-\;i * ** * * *** * * * * ****** * , * , * n^
Ull aao ■ ■ +■ .+■■ ■ 96
Refineries and Distribution ... .*+. * * * *** * m . .. .$++ 98
Natural Gas 99
Biomass Fuels ... .* * ^** „**. , w *. *., .**, „ + , ,*, .**, *, $ 100
Biomass Resource Base * *.. * *.. .*. .**. ***+4.*.. 101
Rural Biomass Markets 101
Commercial Biomass Markets 4 . , @ . *.. ., q+ ****c+* 101
charcoal * * ** ** * * * * *** * \co
Biomass Pricing * *. .** Go . *******e* iq2
Electricity * + + * + . *** *.** e< * aoo . *e 103
Electricity Generation 103
Electricity Transmission and Distribution 104
Institutional Issues ..*. * * *„. *.*. . + . * e . **+**.. ■+<§> 105
Electricity Consumption ,_„„,«,,. 106
outlook for I mprovements 106
Box
Box Page
4-A. Refinery Technology * * * 99
Figures
Figure * Page
4-1. Energy Supply Mix in the Developing World and in the United States, 1987 94
4-2. Energy Supply Mix for Selected Regions of the Developing World, 1987 .. +.*.*. 95
4-3. Primacy Energy Production, Consumption, and Exports in the
Developing World, 1985 95
4-4. Oil Consumption by End Use in the Developing World and in the
United States, 1985 * + * ... *.. **. ..* + . q * *Q**+*.+, 97
4-5. Electricity Generation by Fuel Type in Selected Regions of the
Developing World, 1987 * *. 11+ . ** ca * ...***** *+****+ iq3
4-6. Electricity Consumption by Sector in Selected Regions of the
Developing World, 1985 * * *** ... . e *. * + * + m *** 106
Tables
Table Page
4-1. E nergy Supply Mix, 1 987 94
4-2. Oil Consumption by End-Use Sector, 1985 ....,** *..* 97
4-3. Biomass Energy Resources in Selected Developing Countries 101
4-4. Electricity Generation by Fuel, 1987 104
4-5. Electricity Consumption by Sector, 1985 107
Chapter 4
Energy Supplies in the Developing World
Introduction and Summary
This chapter provides an overview of the energy
supply sector in the developing world— the proc-
esses and technologies by which energy is produced,
converted from one form into another, and delivered
to users. The energy supply sector is critical for
economic development for two reasons: first, eco-
nomic growth depends on the availability of reliable
sources of energy; and second, the energy supply
sector absorbs a large share of investment-over 40
percent of the total public investment in some
developing countries, 1 and about 15 percent of
foreign assistance. 2 The energy supply sector also
relies heavily on other resources that can be scarce
in the developing world, including skilled labor and
management, water, and land.
The energy supply sector is critical to economic
development. Economic growth depends on the
availability of reliable sources of energy, and the
energy supply sector absorbs a large share of public
and foreign aid investment.
The developing countries use a wide range of
energy sources. Coal is the largest primary energy
source in developing countries, due mainly to the
coal-based energy sectors of China and India, the
two largest energy consumers in the developing
world. Excluding China and India, oil and electricity
are the mainstays of commercial energy supplies in
developing countries. In rural and poor urban areas,
traditional biomass fuels are the primary energy
source.
Although the developing world as a whole is a net
oil exporter, the exports originate from a relatively
small number of countries. Most of the countries in
the developing world depend heavily on imported
oil, and these imports consume a large fraction of
export earnings.
The commercial energy supply system-coal, oil
and gas, and electricity-requires large amounts of
foreign exchange, skilled labor, and trained manage-
ment. It is characterized by state ownership, in
contrast to supplies of traditional fuels, which are
largely in private hands.
Commercial energy supplies in many countries
are unreliable and of poor quality due to operational
inefficiencies, rapid increases in energy demand,
problems of reaching dispersed populations served
by inadequate transport systems, and inappropriate
pricing and allocation systems.
Operational efficiencies in the energy industry are
somewhat lower in many developing countries than
in the industrial countries. This record of "poor"
performance often reflects the older equipment and
difficult conditions under which energy facilities
operate. The existence of such differences between
operational efficiencies in reasonably standardized
operations suggests that improvements are possible
if some of these obstacles can be overcome.
Generalizations about energy supplies in develop-
ing countries obscure the heterogeneity of the
developing world. Performance standards vary con-
siderably between countries, as do prospects for
expanding energy supplies in the future.
The Overall Primary Fuel Mix
According to the International Energy Agency
(IEA), coal supplied 35 percent of the developing
world's primary energy 3 in 1987 (table 4-1 and
figure 4-1), followed by oil (31 percent), biomass
fuels (19 percent), primary electricity (mostly hydro-
power) (8 percent), and natural gas (7 percent). Some
analysts believe instead that biomass is in fact the
largest source of energy, supplying up to one-third
of primary energy in the developing world.
iM.Munasinghe, £/e«ric Power Economics (London: Butterworths,1990).
World Bank, Annual Report 1989 (Washington, DC: 1989). Data include ordy International Development Authority (IDA) and International Bank
for Reconstruction and Development (IBRD) lending in fiscal year 1989.
3 ' 'Primary energy" refers to freis in tneii raw state, before they are processed into forms suitable for use by final consum ers. Primary fuels include
coal, oil, gas, biomass, and electricity generated from nuclear,hydro, geothermal, and solar sources. "Final energy," suitable for end-use consumption,
includes electricity generated from fossil fuels as well as primary electricity. For countries with fossil fuel electricity generation facilities, the amount
of electricity in thefinal energy mix is therefore higher than in the primary energy mix, and the amounts of fossil fuels are lower by the amounts used
to generate electricity. Electricity generated from fossil fuels is not included in primary energy in order to avoid double counting.
-93-
94 . Energy in Developing Countries
Table 4-1 — Energy Supply Mix, 1987 (percent)
Fuel
China
India
Brazil
Rest of
developing
world
Total
developing
world
United
States
coal 70
Oil ' 17
Natural gas 2
Other 4
Biomass fuels" 7
Total (percent) 100
Total (exajoules) 26.7
38
6
17
35
24
22
38
41
31
41
3
2
12
7
22
5
26
8
8
9
33
28
22
19
4
100
100
100
100
100
9.6
6.9
42.0
85.2
77.9
a Includes hydropower, nuclear, geothermal, solar, and electricity produced from biomass. Converted at heat equivalent.
b Note that these estimates for the share of energy supplied by biomass are lower than that indicated by detailed field surveys. See, for example, figure 3-1
and app. 3-A.
NOTE: Total may not add to 100 percent due to rounding.
SOURCE: International Energy Agency, World Energy Statistics and Balances 1971 -1987 (Pans: OECD, 1989); IEA, Energy Balances of OECD Countries
1987-1988 (Paris: OECD, 1990).
Figure 4-I — Energy Supply Mix in the Developing World and in the United States, 1987
(exajoules)
Coal
Bi Biomass
16.2
Natural gas
6.0
Developing world
Other*
7.1
United States
Biomass
3.1
Natural gas
17.1
"Includes hydropower, geothermal, other renewable, and nuclear converted at heat equivalent. Biomass numbers
may be underestimates, see text, and alternative source, figure 1-2.
SOURCE: International Energy Agency, WbrldEnergy Statistics and Balances 1971- 1987 (Paris:OECD, 1989); IEA,
Energy Balances of OECD Countries 1987-1988 (Paris: OECD, 1990).
The relative shares of these energy sources in the
overall energy supply mix vary significantly across
different regions and countries, due in part to
unequal endowments of energy resources. Coal
supplies about half of the energy requirements for
developing countries in Asia, due largely to high
levels of coal consumption in China and lndia 4 (table
4-1 and figure 4-2). Oil is the major source of
commercial primary energy for most countries of the
developing world, India and China being the notable
exceptions. Natural gas supplies a relatively small
fraction of energy in the developing world, although
in countries with well -developed resources, gas
often represents an important source of energy (in
Mexico, for example, natural gas supplies 20 percent
of the primary energy consumed). 5 Biomass fuels
are a significant energy source throughout the
developing world, particularly in rural areas and in
the poorest countries.
Overall, the developing world produces more
energy than it consumes, and significant amounts of
both oil and gas are exported from developing
countries (figure 4-3). There are, however, major
disparities among countries: only a few developing
countries export energy, and most import over 50
percent of the commercial energy they consume.
"•International Energy Agency @A), World Energy Statistics and Balances 1971-1987 (Paris: OECD, 1989), using 1987 data. This includes IEA
estimates for biomass consumption and converts electricity to energy units at the heat equivalent,
3 Ibid.
Chapter 4-Energy Supplies in the Developing World • 95
Figure 4-2 — Energy Supply Mix for Selected Regions
of the Developing World, 1987
Exajouies
50
40
30
20
I II
_
- Biomass
= Other*
- Natural ga
- O i I
!ZZZ2 Coal
China
India
Brazil
Rest of
developing world
'Includes hydropower, geothermal, other renewable, and nuclear con-
verted at best equivalent. Biomass numbers maybe underestimates, see
text.
SOURCE: International Energy Agency ,Worfd Energy Statistics and
Balances 1971 -1987 (Pans: OECD, 1989).
Coal
Coal production and consumption in the develop-
ing world are concentrated in a relatively small
number of countries. China, India, South Africa,
South Korea, and North Korea are responsible for 96
percent of the total coal production in the developing
world. China alone accounts for 65 percent of
developing world coal production and 27 percent of
world coal production. 6
The largest consumer of coal in the developing
world is industry, which accounts for over 40
percent of total developing world coal use. 7 The bulk
of the industrial coal use is in China. Other sig-
nificant coal-consuming sectors are electric utilities,
transportation (coal-burning locomotives), and the
residential sector in China, where coal is used for
cooking and space heating.
Rates of coal production are growing rapidly.
Hard coal production in Asia grew at an average
Figure 4-3-Primary Energy Production,
Consumption, and Exports in the
Developing World, 1985
40
Total production (exajouies)
30
20
10
~ Net exports
ISHII! Consumption
Coal
Oil
Natural gas
SOURCE: International Energy Agency„lVorid Energy Statistics and
Balances 1971 -1987 (Paris: OECD, 1989).
annual rate of 5.7 percent from 1980 to 1987, and
lignite production grew at 8.9 percent during the
same period.'This growth is expected to continue.
Coal quality in the less developed world varies
widely, both within and among countries. 10 In India
and China, most coal has a relatively low sulfur
content but a high ash content. "There are few
washing and sorting processes at coal mines, so the
quality of coal supplies is unpredictable. In China,
for example, less than one-fifth of all coal mined is
cleaned before combustion. The rest is used raw,
limiting the efficiency of combustion.
The coal industry accounts for about 16 percent of
total commercial energy investment requirements in
low-income developing countries. 12 In India and
China, the largest coal users, domestic sources
finance most of this investment.
Government-owned entities are responsible for
most coal mining, transport, and distribution in
developing countries. In China, the Ministry of Coal
6 Ibid., p. 59 > 1987 production of hard coal only.
Tfoid., 1985 data. The United States, in contrast, uses most of its coal for electricity generation.
8 As shown in figure 4-3, net coal exports are insignificant, so for the developing world as a whole coal production is the same as coal consumption.
Some individual developing countries do trade incoal — e.g., South Africa is a coal exporter and the Republic of Korea is a coal importer.
9 TEA, op. cit., footnote 4.
I0 The important attributes f coal quality are energy density (typically measured in Joules per kilogram or Btu per pound), sulfur content, and ash
content.
n Tata Energy Research Institute, TERI Energy Data Directory and Yearbook 1988 (New Delhi, India: 1989); Vaclav Smil, "China's Energy,"
contractor report prepared for theOffice of Technology Assessment, 1990.
,2 World Bank, The Energy TransitionK^eveloping Countries (Washington, 1X2: 1983), p. 68. For period 1982-92.
96 •Energy in Developing Countries
Industry controls about 600 of China's 20,000
mines, grouped under 84 Coal Mine Administrations
or Coal Industry Companies. The remaining, mostly
smaller mines are run through local governments at
the provincial, county, or prefecture level, or as
collective township and village enterprises. In India,
the government-owned Coal India Corp. accounts
for 87 percent of production. Coal I ndia, with
675,000 employees, claims to be the world's largest
single corporate employer. 13 The allocation of sup-
plies is also in the hands of a government entity.
Although coal mining technologies in the devel-
oping world are diverse, the major coal producer
-China-relies heavily on manual labor. "About
two-thirds of the extraction from large mines in
China depends on manual labor, as does virtually all
production from locally run mines and small private
pits. Not surprisingly, labor productivities are very
low, averaging less than 1 ton per miner per shift.
This rate is significantly lower than current U.S.
rates, which average about 10 tons per miner per
shift.15
Although China has considerable experience with
a wide variety of advanced underground mining
techniques and has the ability to produce most of the
machinery required, the country does not have
sufficient capital or technical expertise to modernize
its coal industry completely. However, surface
extraction methods, which can be less expensive, are
being used at many newly developed sites. Five
large pits are now under development in China with
a total initial capacity of 50 million tons per year.
The largest of these should eventually produce up to
60 million tons per year.
Transportation requirements often limit coal pro-
duction. In China, coal accounts for 40 percent of all
freight movement, most of which is by rail. China's
already overloaded transport system is struggling to
keep up with its growing coal production. Incom-
plete and poorly configured networks, backups at
mode transfer points, and breakdowns all contribute
to the unreliability of the coal transport system. As
a result, coal-using industries must stockpile up to a
year's supply, or turn to other fuels. Similar prob-
lems occur in India. 16
Oil
Petroleum products are easy to transport and
versatile in use in all sectors and at all scales of
operation; consequently, they play an important role
in the energy sectors of developing countries. These
attributes led to an average annual growth rate of 4.5
percent for oil consumption in the developing world
from 1971 through 1987. "Oil consumption is
expected to continue rising by about 3 percent per
year, thereby doubling between 1985 and 2010. 18
More than one-third of the oil consumed in the
developing world is used for transportation (figure
4-4 and table 4-2). The share of oil used for
transportation varies from 13 percent in China to 42
percent in Latin America-considerably lower than
the 62 percent share in the United States. The
developing world, compared to the United States,
uses proportionally more oil for electricity genera-
tion and for industry. The entire developing world
consumes about 25 percent less oil than the United
States alone.
The bulk of developing world oil production is
concentrated in a few countries— 14 developing
countries account for over 90 percent of developing
world oil production. "Although the developing
world as a whole is a net oil exporter, the exports
originate from a relatively small number of coun-
tries. 20 Most developing countries depend heavily
on oil imports. More than half of the low- and lower
middle-income countries import 90 percent or more
"IDEA, Inc., "Clean Coal Technologies for Developing Countries," contractor report prepared for the Office of TechnologyAssessment, May 1990.
14 This discussion is drawn from Vaclav Smil, "China's Energy," op. cit., footnote 1 1.
15 Relative '• the United States, labor y$ cheaper and mechanization is more expensive— so one would expect greater use of labor inputs and less
mechanization in China than in the United States.
16 For example, one textile mill in India Is converting from coal t0 nee husks as a boiler fuel because of the extreme unreliability Of coal supplies (V.
Kothari, consultant, Isotem Services, New Delhi, India, personal communication, April 1990).
17JJEA, op. cit., footnote 4.
,8 Alan S. Manne and Leo Schrattenholzer, international Energy Workshop: Overview of Poll Responses (Palo Alto, CA: Stanford University, July
1989) International Energy Project.
19 These are, in order of decreasing oil production Mexico, china, Venezuela, Indonesia, Nigeria, Algeria, Libya, Egypt, Brazil, India, Malaysia,
Argentina, Colombia, and Angola. IEA, op. cit., footnote 4.
^ive countries-Mexico, Nigeria, Venezuela, Libya, and Indonesia-account for over 60 percent of LDC oil exports. IEA, op. cit., footnote 4.
Chapter 4~Energy Supplies in the Developing World • 97
Figure 4-4-Oil Consumption by End Use in the Developing World and in the United States, 1985 (exajoules)
nd-try
Industry
5.6
Transportation
8.4
Transportation
18.5
Electricity y
1.2
lllill? Other**
4.8
Transformation*
Developing world
Transformation*
1.2
United States
•Refinery use and losses.
• 'Residential, agricultural, and nonenergy.
SOURCE: International Energy Agency, World Energy Statistics and Balances 1971 -1987 (Paris: OECD, 1989) IEA, Energy Balances ofOECD Countries
1970/1985 (Pans: OECD, 1987).
Table 4-2-Oil Consumption by End-Use Sector, 1985 (percent)
Total
Latin developing
Sector Asia" China Africa America world United States
Electricity 12 15 13 12 13 4
Transformation" 10 10 5 10 10 4
Industry 24 40 18 16 23 14
Transportation 35 13 41 42 35 62
Other" 16 19 17 16 16 11
Nonenergy" 3 4 6 4 4 6
Total (percent) 100 100 100 100 100 100
Total (exajoules) 7.5 3.9 3.5 9.1 24.0 29.9
a excluding China.
transformation includes losses, refinery use, and statistical differences.
c Other is largely residential and agriculture.
d Non-energy use includes waxes, asphalt, and lubricants.
NOTE: Totals may not add to 100 percent due to rounding.
SOURCE: International Energy Agency, World Energy Statistics and Balances 1971 -1987 (Paris: OECD, 1989); IEA, Energy Balances of OECD Countries
1970/1985 (Paris: OECD, 1987).
of their commercial energy; almost all of the imports
are in the form of oil (see app. 1A). By comparison,
the United States imports about 17 percent (net) of
its energy, of which 90 percent is in the form of oil
or oil products. 21 Th e continued expansion Of oil
consumption imposes an economic burden on devel-
oping countries, either in the form of foreign
exchange for imported oil or for investment in oil
exploration and development. Investment in the oil
supply infrastructure is comparable with that in the
electricity sector, each accounting for about 40
percent of total commercial energy investment in the
developing world. 22
In most developing counties, natural resources
such as oil and gas are considered state property.
Ownership or other economic rights to these natural
resources are not bought and sold with the surface
rights (as in the United States), but are reserved for
the state to exploit or to contract out to third parties.
Inmost oil-exporting developing countries, supreme
authority over oil and gas is legally vested in a
central government entity-(e.g., the Ministry of Oil
in Egypt or the Ministry of Petroleum and Minerals
in Oman), which delegates sectoral oversight and
operational responsibility to a national oil company
(e.g., the Egyptian General Petroleum Corp.). In
21 Energy Information Administration (EIA), Monthly Energy Review December 1989, DOE/EIA-O035(89/12) (Washington, DC: Us. Government
Printing Office, February 1990).
^World Bank, op. cit., footnote 12, p. 68.
98 • Energy in Developing Countries
practice, however, there are wide disparities among
different countries in the actual relationships among
executive power, sectoral ministries, and national
operating entities. In oil-exporting countries, such as
Argentina and Algeria, the national operating com-
panies have more power than the ministries they
depend on, due to the companies' autonomous
sources of revenue from significant oil production.
In contrast, in oil-importing countries, such as
Sudan, the national oil companies have little power.
Only the wealthier oil-exporting developing coun-
tries invest much of their own money in exploration
and development. All the poorer ones, and the
majority of the others, rely on the international oil
industry for most of the required capital and
technical expertise. The "enabling mechanism"
under which foreign oil companies invest in devel-
oping countries varies widely from country to
country, but certain aspects are widespread. Gener-
ally, the ministry in charge of the sector authorizes
the national oil company to negotiate exploration
rights with foreign companies. This negotiation may
require the formation of a formal joint venture with
the national company. More typically, however, a
loose relationship is formed in which the two
associates remain entirely separate, with the foreign
party providing all the capital and most of the
technical control of operations. Once a deal is
concluded, it is ratified at the central government
level.
The fiscal arrangements made under these ena-
bling mechanisms have in the past been inflexible,
providing for the same overall rate of government
take on all sizes of fields. These arrangements are
satisfactory for the middle range of expected re-
serves and costs. When oil fields are larger than
expected, however, the government fails to capture
a reasonable share of the profits on the petroleum
being produced. When fields turn out smaller then
expected-a typical occurrence in many developing
countries-the government share precludes eco-
nomic development by the foreign operator. This
discourages the development of small fields. A
further discouragement to foreign oil companies for
the development of oil fields for domestic consump-
tion rather than export is uncertainty over the
availability of foreign exchange for the remittance of
profits.
Despite these disadvantages, U.S. oil companies
in recent years have increased their investments
outside North America, particularly in developing
countries. The prospect of adding to corporate
reserves at half the cost of domestic exploration 23
apparently more than compensates for the uncer-
tainty of foreign investment.
Refineries and Distribution
Although the world's oil refining capacity is
concentrated in the industrialized countries, many of
the developing countries, including oil-importing
countries, have considerable domestic refining ca-
pacity. Over half of the countries in Africa have
refineries, 2i many of which are hydroskimmers (see
box 4~A). In the developing world, refineries are
usually owned by the government, which acts
through a national oil company, although foreign
companies sometimes manage and operate the
refineries.
The refinery product mix in developing countries
differs significantly from that in the United States,
due in part to the inflexibility of the hydroskimming
technology. In developing countries, between 60
and 70 percent of refinery output is diesel and
residual, compared with 30 percent in the United
States. On the other hand, gasoline accounts for
about half of U.S. production, compared with about
20 percent in developing countries. 25 In general,
refineries in developing countries have higher pro-
duction costs than those in the industrialized world:
average refinery operating costs in Africa are $2 per
barrel, compared to $0.75 per barrel in the rest of the
world. 26 Refineries in the developing world also
suffer from large losses due to the use of old,
inefficient technology, as well as poor maintenance.
Refinery losses, which should not exceed 1 percent
^The cost of finding a barrel of oil in the United States is $6, compared with $3 in the rest of the world (Solomon Brothers Inc., Proved Petroleum
Reserves of 30 Large Energy Companies 1980-87 (New York, NY: 1 988), p. 1 4.
^United Nations, 1986 Energy Statistics Yearbook (New York, NY: 1988).
^Ibid.
26j. Gorton, "Oil and Gas Development in Third World Countries, 'draft contractor report prepared for the Office of Technology Assessment April
1990.
Chapter 4™Energy Supplies in the Developing World • 99
Box 4-A — Refinery Technology
Crude oil is a mixture of hydrocarbons. In the
simplest refining process, primary distillation (also
known as hydroskimming), crude oil is simply
heated. This causes the lightest hydrocarbons, such
as gasoline and liquid petroleum gas, to boil off
first. As the temperature increases, different prod-
ucts (such as kerosene and diesel oil) boil off. The
residual, known as residual fuel oil, remains. This
relatively simple process does not allow for much
flexibility in the mix of yielded products. For
example, 18 to 23 percent gasoline and 30 to 55
percent residual fuel oil are typical yields from
primary distillation. 1
Secondary conversion, which includes fluid cata-
lytic cracking (FCC) and hydrocracking, uses high
temperatures to "crack" large molecules into
smaller ones. This process allows for greater
flexibility in product mix, and is often used to
increase the proportions of lighter hydrocarbons,
such as gasoline and LPG, while reducing the
proportion of residual fuel oil. For example, hydro-
cracking allows for up to 86 percent (by weight)
gasoline production, compared with 18 to 23
percent for primary distillation. 'However, second-
ary cracking is relatively expensive and complex.
^.Wijetilleke and A. Ody, "World Refinery Industry— The
Need for Restructuring," World Bank Technical Paper No. 32,
Washington DC, 1984, p. 32a.
2 Ibid, p. 33a.
in a properly maintained and operated refinery, often
exceed 2 or even 4 percent in developing countries .27
for gasoline at the top end and residual fuel at the
bottom. Developing country refineries, which typi-
cally do not have secondary conversion technology,
could not adjust to these changes. As a result, some
of their surplus petroleum products had to be
exported, often at distress prices, while other prod-
ucts had to be imported. 28
Because of these developments in the interna-
tional petroleum market, several developing coun-
tries are producing refined petroleum products at
costs higher than those prevailing on the interna-
tional market. Up until recently, for example,
Liberia had only one refinery, which was poorly
maintained, inefficient, and in need of upgrading.
An economic analysis of this refinery showed that
importing the refined petroleum products directly
and shutting down the inefficient domestic refinery
would result in net savings of $15 million to $20
million per year (U.S. dollars), equivalent to a gain
of about 2 percent in Liberia's gross domestic
product (GDP) .29 Several other African countries are
in similar situations.
The distribution system for petroleum products
plays a key role in determining whether or not the
economy has sufficient and dependable fuel sup-
plies. Because large portions of the population are
scattered in rural areas, and transport infrastructures
are frequently inadequate, the distribution systems
in developing countries are often unreliable. In
addition, price controls, supply monopolies, ration-
ing systems, and requirements for uniform pricing in
all areas may further limit the dependability of
petroleum in developing countries.
During the 1970's and 1980's, both the structure
and the level of petroleum product prices changed.
As the prices of petroleum products increased, coal,
gas, and hydroelectricity were substituted for resid-
ual fuel oil in electricity generation, leading to a
relative decrease in the demand for residual fuel oil.
When the gap between gasoline and diesel prices
widened (due to diesel subsidies and/or gasoline
taxes), consumers switched to diesel cars. As a
result, the structure of demand for petroleum prod-
ucts changed in many countries; the demand for the
middle distillates increased relative to the demand
Natural Gas
Natural gas plays a relatively minor role in the
energy supply system in most developing countries,
supplying only 7 percent of total energy use in the
developing world (see table 4-1). This contrasts
sharply with its role in the industrialized world. In
the United States, for example, natural gas supplies
22 percent of domestic energy consumption. How-
ever, gas is an important source of energy for a small
number of developing countries. Five countries-
Mexico, Argentina, Venezuela, China, and Algeria—
27 Refinery losses result from leakage, evaporation, and spills; they are distinct from "own use" co nsumption, which is oil intentionally consumed
in the refinery process for heat, transport, and drive. See T. Gorton, op. cit., footnote 26.
2«L. Wrjeulleke and A. Ody, "World Refinery Industry-The Need for Restructuring,' World Bank Technical Paper No. 32, Washington+ DC, 1984.
29T. Wilbanks and S. Wright, "Energy for Development: ORNL Returnsto the Third World," ORNL Review, No. 3, 1988.
100 . Energy in Developing Countries
account for 58 percent of developing world natural
gas consumption. Natural gas consumption in the
developing world is concentrated in industry, where
it is used both as a fuel and as a feedstock. Natural
gas is also used for electricity generation.
Discovered gas reserves in many developing
countries are not developed. The current production
levels from developed reserves are only 16 percent
of the level that current proven reserves could
sustain. 30 Thus, many developing countries import
large quantities of crude or fuel oil while possessing
reserves of natural gas that could serve more
economically and with less harm to the environ-
ment. 31
The reasons for the relative underutilization of
natural gas in the developing world lie more in
institutional than in technical or financial con-
straints. Although gas is associated with oil explora-
tion and development and therefore shares many of
the same problems, it has additional difficulties of its
own. In the exploration phase, due to the fiscal or
contractual terms under which gas is discovered, gas
discoveries are often treated as "dry holes" by oil
exploration companies. 32 In the development phase,
gas requires heavy front-end capital investments
(pipelines from producing to consuming regions, or,
in the case of overseas trade, costly facilities and
tankers to liquefy and transport the gas) and long-
term agreements between suppliers and consumers.
And the high cost of building distribution systems in
cities deters the development of domestic markets
for natural gas.
The specific requirements of gas development are
particularly difficult for developing countries, where
financing for large capital-intensive projects is hard
to find, and where the main purchaser for the gas is
the state. When the developer is a private foreign
corporation, the problems are complicated by uncer-
tainty that sufficient foreign exchange will be
available for the investor to repatriate profits. This
last difficulty is especially acute in the case of highly
indebted developing countries where the bulk of
scarce hard currency is earmarked in advance for
payments on debt. Unlike the electricity sector,
which shares many of these characteristics, develop-
ment of natural gas reserves has generally not been
financed by donor agencies.
Biomass Fuels
Biomass fuels are an important source of energy
in the developing countries, supplying over three-
fourths of the total energy consumed in almost all of
the lower income developing countries. 33 The con-
tribution of biomass fuels to total energy supplies in
the entire developing world is unclear. Biomass fuel
consumption is difficult to measure, as much of it
never enters a commercial market. As shown in table
4-1, biomass fuels supply about 19 percent of total
energy according to the I nternational E nergy Agency.
Other researchers, however, estimate this number at
33 to 35 percent. 34
Biomass fuels are the dominant energy source in
rural areas, and they are also widely used in poorer
urban areas, as well as in some large-scale industrial
processes. These 'traditional fuels" consist of
wood (firewood and charcoal), dung (from cattle and
other animals), and crop wastes (e.g., wheat, rice
straw, and sugar cane bagasse). Wood is the most
widely used and preferred fuel due to its superior
combustion characteristics. Its share of total biomass
energy supply varies widely among the developing
countries, according to region and agricultural and
forestry resource base. In many regions of Africa and
Latin America (with the important exception of
Brazil, where bagasse is also used), wood is the
primary form of biomass energy used. I n Asian
countries, wood remains the dominant biomass fuel
(accounting for one-half of all biomass consumed in
30 World Bank, op. cit., foomote 12, p. 36.
3iTo mention @t the sub-Saharan African countries with undeveloped gas fields: Mozambique, Ethiopia, Somalia, Madagascar, Cote d'lvoire,
Equatorial Guinea, Sudan, Senegal, Tanzania, and Namibia.
32]jy one estimate, about half of the natural gas produced in the developing world is flared or otherwise wasted. The comparable number ^ OT Eastern
Europe is 2 percent. A. Mashayekhi, "Natural Gas Supply and Demandin Less Developed Countries," Annual Review of Energy, vol. 13, 1988, pp.
119-129.
33 The role of biomass fuels in the total energy supply varies greatly among countries, show @ generally an inverse relationship with GNP per capita.
For example, Ethiopia (GNPper capita = $U.S. 130) meets 92 percent of its energy needs with biomass fuels, while Argentina(GNPpercapita=$U.S.
2,390) meets only 7 percent of its needs with biomass fuels.
M K. Smith, "The Biofuel Transition," /"aciyic and Asian Journal ofEnergy, 1987, pp. 13-32; P. O'Keefe, J. Soussan,B.Munslow,andD.Spence,
"Wood Energy in Eastern and Southern Africa," Annual Review ofEnergy, vol. 14, 1989, pp. 445-468.
Chapter 4~Energy Supplies in the Developing World .101
China, and three-quarters in India), but crop wastes
and animal dung also play a significant role. 35
Biomass Resource Base
For a variety of reasons, the fuelwood supply base
i s shrinking rapidly. This could have serious impacts
on the populations that depend on forests for fuel,
food, fiber, fodder, and other needs.
The total global annual growth of forest biomass
is subject to great uncertainty, but has been esti-
mated to be about 50 times annual wood consump-
tion or five times total annual energy consumption,
including fossil fuels. Despite this apparently large
average global supply, there are acute and growing
shortages of fuelwood both locally and regionally.
Some regions, such as Asia, have very little forest
stock per capita (table 4-3). Within regions, some
countries are well endowed with biomass energy
resources, while others have totally inadequate
supplies; and within countries themselves, local
abundances and shortages are common. Zaire, for
example, consumes only 2 percent of its sustainable
yield of forest biomass but has serious deforestation
around Kinshasa. 36
Despite the uncertainties regarding rates of bio-
mass energy use and supply, it is clear that the
populations affected by fuelwood shortages are
increasing. The United Nations Food and Agricul-
ture Organization (UNFAO) has estimated that the
number of people suffering acute shortages of
fuelwood will increase from about 100 million in
1980 to over 350 million in the year 2000. Such
shortages increase fuel costs for urban dwellers,
lengthen the time spent foraging for fuel by rural
dwellers, and rob the soil of nutrients as people
switch from wood to crop wastes and dung (although
the impact of this nutrient loss maybe limited except
in the much longer term).
Rural Biomass Markets
Much of the biomass fuel supply in developing
countries-especial ly twigs, branches, dung, and
crop wastes— is gathered locally and used by family
members without entering commercialized markets.
Table 4-3-Biomass Energy Resources in Selected
Developing Countries
Sustainable energy yield
(GJ per capita per year
Crop Animal
Country Wood residues dung
Congo 570 1 ~~
Brazil 350 8 16
Zaire 135 1 1
Argentina 123 25 40
Thailand 37 9 4
Nepal 21 7 13
Burkina Faso 10 5 7
India 7 5 6
Bangladesh 2 4 5
China — 7 3
— data not available or not applicable
SOURCES: G. Barnard and L. Kristofferson, Agricultural Residues as Fuel
in theThird World (London: Earthscan, 1985); D. Hall, G.
Barnard, and P. Moss, Biomass for Energy in the Developing
Countries, Pergamon Press, 1982. R. Moss and W. Morgan,
Fuelwood and Rural Energy Production and Supply in the
Humid Tropics, (Dublin, Ireland: Tycooly International Publish-
ing Ltd, 1981).
These supplies are gathered free of charge (if the
considerable cost of the labor used in gathering is not
included) from fields, hedgerows, gardens, and
nearby forest lands. In some cases, however, the
poor may have to "pay" with labor services for the
privilege of gathering biomass fuels from privately
owned land.
Commercial Biomass Markets
Biomass fuels, notably logs and charcoal, are also
traded in commercial markets far from their origin in
government and private forests, farms, or planta-
tions. Low-income urban households and small
commercial enterprises use the bulk of these fuels.
In some cases, however, biomass fuels are used for
advanced industrial applications, as in the case of
charcoal for iron smelting in Brazil. In such cases,
the industrial users often organize the biomass fuel
supplies.
Unlike other forms of energy, supplies of com-
mercialized biomass fuels are largely in the hands of
the private sector. Much of the fuelwood may be
grown on privately owned land, 37 and the transport
and distribution charnels for commercialized bio-
35 Crop wastes account for one-half of total traditional energy supply in China and just over 10 percent in India. Animal ,j un g accounts for about 20
percent of traditional fuel use in India.Pakistan, and Bangladesh but under 2 percent in China.
3*R, Moss and W. Morgan, Fuelwood and Rural Energy Production and Supply in the Humid Tropics (Dublin, Ireland: Tycooly International
Publishing Ltd., 1981).
37 In several Indian cities, for example, government lands were found to provide less thanlO percent of total fuelwood supplies (MAlam, J. Dunkerley,
and A. Reddy, "Fuelwood Use in the Cities of the Developing World: Two Case Studies From Ire@" Natural Resources Forum, vol. 9, No. 3, 1985).
102 •Energy in Developing Countries
mass, fuels are typically in private hands, as are
charcoal kilns. 38
On the other hand, the fuelwood trade is often
subject to government regulation, with strict rules
about cutting trees in government forests and even
on private lands. Although it is believed that such
regulations are not strictly enforced, often because
of the difficulty of enforcement, proscriptions against
cutting trees can discourage the development of
long-term supplies, as farmers and others are unwill-
ing to invest in tree planting for fuelwood if they
have no assurance that they can harvest the trees at
maturity. Fuelwood prices may also be subject to
price controls (in Senegal, for example, charcoal
prices are controlled by the government). And in
some cases, governments may play a role in the
distribution system as well. Compared with com-
mercial fuels, however, the biomass trade is rela-
tively unregulated.
The transport of wood and charcoal to urban areas
is carried out in a variety of ways. In India, poor
women carry head loads of fuel to urban markets; in
Niger, camels carry fuel into the capital city of
Niamey; and elsewhere fuel is carried by bicycle,
animal cart, moped, and other means. In higher
income areas, trucks or trains carry the bulk of the
fuel.
Charcoal
In rural areas, the cutting of fuelwood and its
conversion to charcoal is a major source of income
and nonagricultural employment. Charcoal is made
by stacking the wood, covering it with a layer of dirt,
and letting it burn with a limited supply of air. The
efficiency of converting wood to charcoal in these
simple earthen kilns is quite low, typically ranging
from 40 to 60 percent. 39 If a capital investment is
made, ranging from a few hundred dollars for simple
modifications to traditional kilns 40 to $100,000 or
more for a modern continuous retort, higher energy
efficiencies can be achieved.
Although it is widely believed that charcoal is
cheaper to transport than wood due to its higher
energy content by weight, detailed studies have
found that the transportation costs for wood and
charcoal are about the same. 41 The higher energy
content of charcoal per unit weight is counter-
balanced by its lower weight per unit volume.
Despite its higher price, charcoal is widely used in
some countries, particularly in urban areas where
people have cash incomes. A 1970 report from
Thailand, for example, indicated that 90 percent of
the wood cut for urban markets was converted to
charcoal. 42 It has several important advantages over
wood. Charcoal is impervious to insect attack,
unlike some wood species that must be used within
as little as a month of drying to avoid significant
losses to termites. 43 As it is nearly smokeless,
charcoal cooking can be done indoors in relative
comfort without blackening walls or metal pots with
soot. In addition, charcoal causes little smoke
irritation to eyes or lungs. Although it can emit large
amounts of dangerous carbon monoxide and other
pollutants, which is a health hazard in poorly
ventilated kitchens, charcoal causes little obvious
discomfort to the user. Additionally, once lit,
charcoal fires need little attention from the cook,
whereas wood fires require frequent adjusting of the
fuel.
Biomass Pricing
When people move from rural to urban areas in
developing countries, they typically continue to
follow traditional patterns of biomass fuel use. In
contrast to the labor-intensive collection of biomass
fuels in rural areas, however, the urban poor often
have no choice but to purchase fuelwood or charcoal
in commercial markets. In Tanzania, the cost of
purchasing these fuels reportedly ranges as high as
38 M. Alam, J ■ Dunkerley, and A. Reddy, Ibid.
^Charcoal Production Improvement for Rural Development in Thailand (Bangkok, ThaUqji R yal Thai Government and U .S. Agency for
International Development, 1984); DBarl, CharcoalProduction, Sudan Renewable Energy Project, Energy Research Council, USAID, Report No. 002,
Khartoum, Sudan, February 1984; J. Wartluft and S. White,Comparing Simple Charcoal Production Technologies for the Caribbean (Arlington, VA:
Volunteers in Technical Assistance, 1984).
*°K, Christophersen, G. Karen, and J . Seve, "Production and Transportation olFuelwood and Charcoal From Wood Surplus t (Deficit Regions in
Niger: Technical and Economic Feasibility" (Washington DC: Energy/Development International, March 1988).
4lx Wood and S. Baldwin, ' 'Fuelwood and Charcoal Use in Developing Countries, " Annual Review of Energy, vol. 10, 1985, pp. 407-429.
42j Arnold, "Wood Energy and Rural Communities, " Natural Resources Forum, vol. 3, 1979, pp. 229-252.
43 Simon Nkonoki and Bent Sorensen, ' 'A Natural Energy Study m Tanzania: The Case of Bundilya Village,' ' Natural Resource Forum, vol. 8, No.
1, 1984, pp. 51-62.
Chapter 4— Energy Supplies in the Developing World . 103
40 percent of the income of poor families. 44 More
typically, energy accounts for 5 to 10 percent of the
expenditures of poor households. 45
Biomass fuel prices in urban markets often rise
rapidly as wood resources are seriously depleted,
and then closely follow fossil fuel markets. Biomass
costs cannot rise very far above the cost of an
equivalent amount of useful energy from kerosene or
LPG, as users can and will then switch fuels/
Families that purchase modern stoves and fuels,
however, rarely discard the older stoves. Maintain-
ing both technologies allows easy and flexible
switching between fuels in response to availability
and price. Following the 1973 and 1979 oil price
increases, for example, many people switched back
to wood and charcoal for their cooking needs. In
Malawi the use of kerosene, primarily for cooking
and lighting, declined by 24 percent between 1973
and 1976. 47
Electricity
Electrification plays a central role in promoting
economic and social development in any nation. At
the same time, the electricity sector consumes large
amounts of economic, social, and environmental
resources. Accordingly, the electric power sector
receives significant attention and resources from
both developing country governments and interna-
tional development agencies. For example, the
World Bank directs over 80 percent of its energy
lending to the electricity sector.
Although electricity accounts for less than 9
percent of the energy used by consumers in develop-
ing countries, 48 electricity production in the devel-
oping world is increasing rapidly, at an average
annual rate of 7.6 percent. 49 However, this rapid
growth still leaves the developing world at a far
lower level of electricity production than the indus-
trialized world: average annual electricity produc-
tion in the developing world is about 520 kilowatt-
Figure 4-5--Electricity Generation by Fuel Type in
Selected Regions of the Developing World, 1987
Thousand GWh
m Hydro/other*
= Nuclear
m Natural Gas
U Oil
- Coal
Africa Latin America India China
'Includes hydropower, geothermal, and other renewable.
SOURCE: International Energy Agency, World Energy Statistics and
Balances 1971-1987 (Paris: OECD, 1989).
hours (kWh) per capita, 50 as compared with the U.S.
figure of 10,500 kWh per capita. 51
Electricity Generation
In the developing world, as in the United States,
a variety of fuels and technologies are used for
electricity generation (see figure 4-5 and table 4-4).
Coal use for electricity production in the develop-
ing world is concentrated, largely in countries with
sizable domestic coal reserves (India, China, and
South Africa). Similarly, natural gas generating
capacity is found mainly in the few countries where
natural gas is produced domestically. In other
countries, oil is often used for electricity generation.
Hydroelectric facilities range from microhydro-
power stations with less than 0.1 megawatt (MW) of
capacity to large-scale hydropower plants such as
the 12,600 MW Itaipu facility in Brazil. For coun-
tries with hydroelectric potential, hydropower offers
an indigenous source of electricity generation with
very low operating costs, although the capital costs
^E.M.Mnzava," Village Industries vs. Savannah Forests," UNASYLVA, vol. 33, No. 131, 1981, pp. 24-29.
45 Gerald Leach and MarciaGowen, "Household Energy Handbook," World Bank Technical Paper No. 67, 1987,p.50.
^Douglas F. Barnes, "Understanding Fuelwood Prices in Developing Nations," World Bank, Household Energy Unit, Industry and Energy
Department, Oct. 31, 1989.
47j. Arnold, "Wood Energy and Rural Communities, " Natural Resources Forum, vol. 3, 1979, pp. 229-252.
""IEA, op. cit., footnote 4, pp. 112, 120, 124, 128.
49 For the period 1971-87. IEA, op. cit., footnote 4.
S0 U.S. Agency for International Development PowerShortages in Developing Countries: Magnitude, Impacts, Solutions and the Role of the Private
Sector (Washington, DC: Office of Energy, U.S. AID, March 1 988), p. 2.
5I Energy Information Administration, Annual Energy Review, DOE/EIA-0384(88) (Washington, DC: U.S. Government Printing Office, 1989).
104 • Energy in Developing Countries
Table 4-4-Electricity Generation by Fuel, 1987 (percent by kWh delivered)
Fuel Africa Latin America China India
United States
coal 52 3 68 66 57
lil 14 24 12 8 5
Natural gas 14 8 1 11
Nuclear 3 1 2 18
Hydro/other" 18 63 20 23 10
Total (percent) 100 100 100 100 100
Total(GWh) 283,340 520,290 497,320 217,500 2,732,530
'Less than l%.
a Includes hydro-power, geothermal, and other renewables.
NOTE: Does not include heat losses. Totals may not add to IOO percent due to rounding.
SOURCE: International Energy Aasncv.Wforid&je/gy Statistics and Balances 1971-1987 (Paris: OECD.1 989); \BKJEnergv Balances of OECD Countries
19 8 7-1988 (Paris: OECD, 1990).
are high. In Latin America, hydropower plays a large energy to electricity, wind-driven generators, and
role, supplying almost two-thirds of total electric- the burning of waste material. Although these
ity. 52 More than 90 percent of Brazil's total electric- sources contribute relatively small amounts to total
ity supply comes from hydropower. 53 In Africa, electricity supplies in the developing world, there is
almost half of the electricity generated outside of strong policy interest in expanding the use of these
South Africa comes from hydropower, 54 and more alternative generating technologies,
than two-thirds of the countries in Africa produce
electricity from hydroelectric plants. 55 The operating efficiencies of electricity gener-
ating plants are generally lower in developing
Only eight developing countries were generating ™ ntries *!" in the UnitedStates. 59 Although many
electr city from nuclei powerplants as of late actors a *f P° w h er Plant efficiency (notably input
inon56 J . i u i 4. i j fue qua ity), the use of ess efficient, poor y
1989 although several more have plants planned stained technologies contributes to the low
or under construction Nuclear power generation ^ l6en6es of developing world powerplants.
involves high capital costs, very large scales of
operation, and considerable technical expertise. In
two upper income developing countries, Taiwan and Electricity Transmission and Distribution
Korea, nuclear power provides over 40 percent of niecinciiy iransmission ana uistrwunon
total electricity generation. In ether countries, how- Transmission and distribution systems in the
ever, there have been some difficulties with nuclear developing world have relatively high losses. A
power-e.g., two of India s three operating nudear recent of 76 developing countries found that,
plants have suffered high outage rates, and thar in one . ha , f of the countries survey ed, transmission
fourth plant has had construction delays of 7 years. and distribution losses (as a share of total genera-
tion) exceeded 15 percent, compared with typical
A number of alternative generation technologies losses of less than 10 percent in the industrialized
have also been used in developing countries, includ- countries. 60 These losses include both technical
ing cogeneration, geothermal, conversion of solar losses and unmetered consumption (theft).
52JEA, °P- c i'-> footnote 4.
53 Ibid.
^United Nations, op. cit., footnote 24.
^Ibid.
^These are Argentina, Brazil, India, Mexico, Pakistan, Republic of Korea, South Africa, and Taiwan. United Nations, ' 'Energy Exploration and
Development Trends in Developing Countries," Report of the Secretary-General, May 14, 1990.
^Argentina, Brazil, China, Cuba, Mexico, India, jm d Republic of Korea all had nuclear powerplants under Construction in 1989. Ibid.
38 Tata Energy Research Institute, op. cit., footnote 1 1 .
^United Nations Conference °. Trade and Development (UNCTAD), Technology Policy in the Energy Sector: Issues, Scope and Options fo r
Developing Countries,VNCVADfTT/90, June 15, 1989.
«°Ibid.
Chapter 4— Energy Supplies in the Developing World • 105
Institutional Issues
Governments in both the developing and industri-
alized world generally have taken leading roles in
directing the development and operation of the
electric power sector; this reflects both the impor-
tance of electric power in meeting economic and
social objectives and the high cost of electric power
systems.
Because electricity is considered an important
tool within a broader national development strat-
egy 61 , developing countries often subsidize elec-
tricity prices. In Pakistan, for example, 60 percent of
the cost of electricity is subsidized; in India, the
figure is 20 percent. 62 Many countries have subsi-
dies for electricity in selected sectors (e.g., agricul-
ture and residential), reflecting either the political
influence of the subsidized sectors or government
interest in promoting certain economic or social
ends. Although electricity prices nearly doubled
between 1980-81 and 1986-87 in India, for example,
current prices are still far lower than supply costs for
residential and agricultural consumers. 63
There may be many reasons for subsidizing
electricity prices in different sectors. For example,
rural electrification is promoted as a means to reduce
migration to cities by improving economic opportu-
nities and lifestyles in rural areas. Supporters argue
that these subsidies benefit society as a whole and
not just rural consumers. Others rogue that there is
no conclusive evidence that rural electrification can
actually produce this benefit. 64 Rural electrification
is also seen as 'a powerful instrument at the disposal
of central governments to foster political stability in
rural areas," although again the evidence is mixed. 65
Subsidized electricity prices also allow electricity
to compete with subsidized prices for alternative
fuels. Similarly, when prices of farm products are
controlled and kept artificially low, electricity prices
(and the prices of other inputs such as fertilizers) are
often subsidized to keep farms operating. 66
Subsidized electricity can have negative impacts,
including contributing to power shortages, since
consumers tend to use more subsidized electricity
than they would otherwise. Moreover, price subsi-
dies keep power company revenues at levels inade-
quate for developing additional supplies and even
for maintaining existing facilities efficiently. Higher
prices, on the other hand, limit supplies to higher
income groups, an outcome that may conflict with
the social goals of electrification.
Capital spending on electricity systems in the
developing world is currently estimated at $50
billion to $60 billion annually. 67 Even at that high
level of expenditure, investment is expected to be
inadequate to meet demand. The United States
Agency for International Development (U.S. AID)
has estimated that meeting the growing demand for
electricity will require capital investment of around
$125 billion per year over the next two decades. 68
This enormous capital mobilization requirement
represents a large fraction of both total economic
activity and total gross domestic investment in the
developing world. The total economic output of all
lower- and middle-income countries as measured by
GDP was $2,716 billion in 1987, with total gross
domestic investment of $662 bill ion. 69 Much of the
capital costs of electricity plants must be paid in
foreign exchange, 70 leading to balance of payment
problems and compounding the problems of high
operating and capital costs in the electric sector.
61 United Nations Center for Human Settlements (Habitat), Guidelines for the Planning of Rural Settlements and Infrastructure: Electrification— A
Methodology (Nairobi, Kenya: United Nations,1985), p. 43.
s*U.S. Agency for International Development, Op. cit., footnote 50, P-26.
a A.Fm^metdi.,ApplicationofDemand-Side Mana g^ mem (DSM)toRelieve Electricity Shortages in India, draft contractor report prepared for
the Office of Technology Assessment, April 1990, p. 59.
**For a discussion of these issues, see D. Barnes, Electric Power for Rural Growth (Boulder, CO: Westview Press, 1987), pp. 109-118.
"United Nations Center for Human Settlements (Habitat), op. cit., footnote 61.
«Mohan Munasinghe, Rural Electrification for Development (Boulder, CO: Westview Press, 1987), P- 247.
67 U.S. Agency for I nternational Development^ cit., footnote 50, p. i v.
M Ibid.,p.25. A slighUylower esnrMt e($60 t -$100 b,ui o n ) fromtoeWorldBank is given jnA. Churchill andR. Saunders, "Financing Of the Energy
Sector in Developing Countries," World Bank, Industry and Energy DepartmenfWorlring Paper Energy Series, Paper No. 14, April 1989.
69 WorldBank, World Development Report 1989 (New York, NY: Oxford University Press, 1989), tables 3 and 9.
70 U.S. AID estimates that around 45 percent of capital investment will be in foreign exchange. U.S. Agency for International Development, footnote
50, p. 25.
33-718
90
106 •Energy in Developing Countries
Even with the enormous financial, technical, and
institutional effort targeted to the development of the
electricity sector over the past decades, capacity still
is often insufficient to provide reliable, high-quality
power in developing countries. Outages are common
in many countries. For example, Bombay experi-
enced 1,000 outages annually over a recent 5-year
period. 71 Even when power is available, voltage
fluctuations are often extreme, restricting the use of
some types of equipment. The electronic circuits of
today's compact fluorescent light bulbs, for exam-
ple, do not tolerate wide voltage fluctuations; 72 and
computer operations are disrupted by outages. As a
result, in many countries, poor power quality and
lack of reliability undermine the economic benefits
of electric power.
Low reliability results in formidable losses in
economic productivity. Load shedding in India is
estimated to cost the equivalent of 1 to 3 percent of
GDP annually. 73 Accurately estimating the produc-
tivity lost when existing equipment cannot be
operated due to power outages is difficult, and
estimating productivity lost as industry forgoes the
purchase and use of new electric equipment is even
more uncertain; however, the impacts may be quite
large.
Ironically, a few developing countries suffer from
an excess of electric capacity. For example, it has
been estimated that seven East African countries
have approximately 7,000 MW of excess generating
capacity (i.e., capacity over and above what is
needed for reliable system operation) .74 This situa-
tion results from the "lumpiness" of electric gener-
ating facilities, especially hydropower. Putting a
large new generating facility in service before the
domestic load can absorb the new supplies results in
overcapacity, and ties up scarce capital. 75
Figure 4-6-Electricity Consumption by Sector in
Selected Regions of the Developing World, 1985
Thousand GWh
300
100
= Other
= Commercial/public
m Residential
= Agricultural
= Industrial
China
India
Brazil
SOURCE: International Energy Agency, World Energy Statistics and
Balances 1971 -1987 (Pans: OECD, 1989).
Electricity Consumption
Industry consumes most of the developing world's
electricity (see figure 4-6 and table 4-5). In contrast,
electricity use in the United States is divided among
the industrial, residential, and public service/
commercial sectors (see table 4-5). China, the largest
electricity user in the developing world, 76 uses 75
percent of its electricity in industry. Similarly, India
and Brazil use over half their electricity in industry.
Agriculture uses large amounts of electricity for
pumping in India and China, while residential
lighting and appliances use large amounts of elec-
tricity in Brazil.
Outlook for Improvements
The mix of energy supplies varies widely in the
developing world— from China's heavy reliance on
coal in the industrial and residential sectors to
Brazil's extensive use of hydropower-based electric-
ity. Despite the diversity of sources, however,
71 Tata Energy Research Institute, Two Strategies for Electric Load Leveling for Inom,Fbi^lFmBl^ep(xt(tfe!wl)eM, India: 1987), p. 6., as cited
inJ.VanDomelen.i'ovcerw Spare: The WorldBank and Electricity Conservation (Washington, DC: World Wildlife Fund andConservation Foundation
1988) p. 4.
72 LawrenceBerkeley Laboratory, Energy Technology for Developing Countries: Issues for the US. National Energy Strategy, LBL-28907 (Berkeley,
CA: December 1989).
73 U.S. Agency for International Development, op. cit., footnote 50.P-21.
T\, Hume, ' 'Energy Efficiency in Developing Countries, " i n M. Munasinghe and R. Saunders (eds.), Energy Issues in the Developing World, World
Bank Industry and Energy Department Working Paper, Energy Series Paper No. 1 (Washington DC: WorrBank, 1988).
75 Excess capacity can also l ea d *• distortions in pricing and demand. For example, the large surplus capacity in Brazil when Itaipu came on-line led
the electricity authorities to offer industry exceptionally strong incentives to buy electricity. Electricity was so cheap that it was used by industrial
customers virtually as a boiler fuel. Within a few years, however, load growth in othesectors reduced surplus capacity, but the industrial users were
by then reluctant to give up the highly favorable rates.
76 0ver one-fourth of the electricity in the developing world is used in China.
Chapter 4™Energy Supplies in the Developing World .107
Table 4-5-Electricity Consumption by Sector, 1985 (percent)
End-use sector China India Brazil United States
Industry 75 59 58 33
Agriculture 12 17 3 1
Residential 7 12 20 35
Public service/commercial 5 8 20 31,
Railroads 2 2 1,
Other/unspecified 1 o
Total (percent) 100 100 100 100
Total generation (GWh) 410,700 188,500 192,700 2,621,900
• Less than 1 percent.
NOTES: Totals may not add to 100 percent due to rounding. Brazil, China, and India account for 48 percent of
developing world electricity consumption.
SOURCES: Tata Energy Research Institute, TERI Energy Data Directory and Yearbook J988(New Delhi, India: 1989);
\EA,Wbrid Energy Statistics and Balances 1970/1 985 (Paris: OECD, 1987); IEA, Energy Balances of
OECD Counties 1987- 1988( Pans: OECD, 1990).
several important characteristics of the energy sup-
ply sector can be identified:
• The technologies in use are typically older, less
efficient, and less sophisticated than compara-
ble technologies in the industrialized countries.
• The electricity sector accounts for a large share
of foreign exchange resources. Rapid growth in
electricity demand and the high capital require-
ments of the sector suggest that the gap be-
tween needed and available capital for elec-
tricity system expansion will widen. Therefore
the electricity systems in many countries could
continue to be relatively undependable, ineffi-
cient, and technologically outdated.
• The public sector plays a dominant role in most
aspects of energy supply, with the exception of
biomass.
These characteristics of the energy supply sector
in the developing world suggest a number of
productive opportunities for improving the effi-
ciency of the energy supply system, recognizing that
there are wide variations among developing coun-
tries, and that many characteristics that appear
inefficient or undesirable when viewed from the
current perspective of highly developed nations may
in fact represent rational choices given prevailing
social, economic, and technical conditions.
A number of institutional, technological, and
engineering options can be considered for improving
the extraction, processing, and conversion of energy
supplies. In oil and gas development, for example,
options include both the deployment of new technol-
ogies, such as horizontal drilling, and the develop-
ment of innovative financing mechanisms. Options
for coal include technologies such as washing and
screening, as well as strategies for minemouth
electricity generation to relieve pressure on transport
systems. Similarly, institutional issues are critical in
any discussion of improving the sustainability of
biomass resources.
Both developing country governments and inter-
national development agencies already pay much
attention to opportunities in the electricity sector.
Technological opportunities range from industrial
cogeneration, to upgrades of transmission and distri-
bution system efficiencies, to the use of more
efficient consumer appliances. Institutional opportu-
nities include the contribution of nonutility genera-
tors to electricity networks.
Although technology can do much to improve the
energy supply sector, other factors also affect its
operation: financial issues, such as subsidies for
electricity prices or the high cost of natural gas
transportation; institutional and management issues,
including shortages of trained personnel; and the
incentive structure, notably the dominant role played
by government in fossil fuel exploration and deliv-
ery systems, all strongly influence system opera-
tions, management, and decisionmaking in the
energy supply sector.
Chapter 5
Energy and the Environment
in Developing Countries
Contents
Page
I ntroduction and Summary Ill
The Rural Sector 113
Causes of Environmental Degradation * *.*.**+...*.* *****..** 113
Environmental Impacts 119
The U rban Sector 123
Causes of Environmental Degradation 123
Environmental Impacts 124
Energy Production and Conversion 128
coal 128
Oil and Gas 128
Hydroelectricity 128
Biomass 129
Nuclear Energy 129
Solar, Wind, and Other Renewable Energy 130
Greenhouse Gases and Global Climate Change . * 130
Government Policies 131
Figures
Figure Page
5-1. Suspended Particulate Matter Levels in Selected Cities, 1980-84 125
5-2. Sulfur Dioxide Levels in Selected Cities, 1980-84 ., ** 126
Tables
Table Page
5-1. Causes and Consequences of Environmental Degradation in Rural Areas 112
5-2. Estimates of Land Use Changes in Closed and Open Tropical Forests,
circa 1980 115
5-3. Changes in Land Area Uses *.* *,*....**.**„..*... .**+**«.... 117
5-4. Sedimentation Rates of Some Reservoirs in India *_***__* * ( * 121
Chapter 5
Energy and the Environment in Developing Countries
Introduction and Summary
Many developing countries are experiencing sig-
nificant environmental degradation. In rural areas,
population pressure and low agricultural productive-
ties are among the factors forcing people into
marginal and ecologically fragile lands. Woodlands
are being cleared for cropland and pastures and are
being commercially logged. The use of biomass for
fuel or fodder places further demands on woodlands
and grasslands, particularly in arid regions with high
population densities. Farming, ranching, logging,
and the use of biomass fuels are all necessary if the
people dependent on these resources are to survive.
But these various pressures can also have negative
impacts: destruction of tropical forests and biodiver-
sity; desertification; soil erosion and increased
downstream flooding and siltation; and air pollution-
local, regional, and global.
In many urban areas of developing countries,
rapid population growth, inadequate infrastructure,
and economic and industrial growth with minimal or
inadequately enforced environmental controls have
led to high levels of pollution. Levels of sulfur
dioxide, particulate, ground-level ozone, and nitro-
gen oxides often exceed those in industrialized
countries. Major sources include electricity genera-
tion, transportation, and industrial production. Greater
use of fossil fuels in the modern, primarily urban,
sector can also lead to environmental degradation
and pollution in the rural areas where these fuels are
extracted from the ground and transported to the
cities, and where hydroelectric facilities are sited.
Many have viewed environmental costs-degra-
dation and pollution of the natural resource base-as
the price that must be paid in order to develop
economically. 'Increasingly, however, others argue
that environmental protection and economic devel-
opment are tightly interconnected and mutually
supportive. 2 The landless peasants who migrate to
fragile watersheds in order to feed their families, for
example, clear land that is highly susceptible to
erosion. This can lead to serious soil loss and
downstream flooding and siltation-ultimately re-
ducing the productivity of their land as well as that
of land downstream. In turn, this can exacerbate their ,
and others' poverty. This may be particularly
significant in developing countries because their
economies are so heavily dependent on agriculture.
For the 2.8 billion people in the lowest income
countries, agriculture accounted for fully 31 percent
of gross domestic product (GDP) in 1987 and an
even greater portion of employment. 3 Economic
analyses that include environmental impacts are
improving knowledge of this interdependent and
may assist policymakers in more wisely making
these tradeoffs.
Energy production and use contribute to environ-
mental degradation in developing countries. Other
contributing factors include population growth,
inequitable land tenure, unsustainable agricultural
and forestry practices, industrialization, and govern-
ment policies. In order to understand the role of
energy in overall environmental degradation, it is
therefore necessary to include these other factors in
the analysis.
Energy, used wisely, might also potentially pro-
vide several important environmental benefits in
developing countries. Greater energy inputs into
agriculture in the form of tractive power, fertilizer,
and irrigation, for example, can substantially im-
prove agricultural productivities where soils and
climates are appropriate, and might help slow the
expansion of agricultural lands necessary to feed a
burgeoning population. (At the same time, however,
modern agriculture might also cause environmental
damage: by overuse of pesticides, herbicides, and
fertilizers; by waterlogging and salinizing irrigated
'Clem Tisdell "Sustainable Development: Differing Perspectives of Ecologists and Economists, and Relevance to UXs," World Development,
vol.16, No. 3, 1988, pp. 373-384.
zWorld Commission on Environment and Development, Our Common Future (New York, NY: Oxford University Press, 1987).
3 WorldBank,Wor/d Development Report 1989 (New York, NY: Oxford University Press, 1989), indicator tables land 3.
-Ill-
112 •Energy in Developing Countries
Table 5-l-Causes and Consequences of Environmental Degradation in Rural Areas
Consequences
Direct cause
Underlying cause
Deforestation
Loss of biodiversity
Soil erosion
Flooding
Shifting agriculture
Permanent agriculture
Permanent pasture
Commercial logging
Commercial agriculture, ranching
Desertification
Air pollution
Use of Biomass Fuels
Use of forest biomass for fodder
Agricultural expansion onto fragile lands
Overgrazing
Burning of grasslands
Use of biomass fuels
Climate change
Use of biomass fuels
Slash and burn agriculture
Burning of grasslands
Salinization and water-logging Poor planning and management
of irrigated lands Inadequate investment in infrastructure
Population growth
Poverty
Lack of land tenure
Low-level agriculture inputs
Mechanization of agriculture and/or the consolida-
tion of agricultural lands
Destructive logging, lack of forest management and
protection, poor reforestation
Increased access to forests along logging roads for
farmers and ranchers
Production for export markets
Fiscal policies and legislation, in part to promote
exports of primary products due to need for
foreign exchange to service debt.
Inappropriate economic valuations of natural re-
sources and biodiversity
Inefficient use of fuelwood; overcutting of fuelwood
resources
Shortages and lack of alternative sources of fodder
Population growth
Poverty
Lack of land tenure
Low-level agriculture and/or the consolidation of
agriculture lands
Inefficient use of fuelwood; overcutting of fuelwood
resources
Various; not well understood
Lack of access to higher quality fuels and stoves
Population growth
Poverty
Lack of land tenure
Low-level agricultural inputs
Cheap or free water contributing to inefficiency
SOURCE: Office of Technology Assessment, 1990.
lands; and by use of these techniques under inappro-
priate soil and climatic conditions). 4
Energy-efficient technologies may offer the po-
tential of simultaneously reducing the total cost of
delivering energy services and cutting pollution.
New industrial processes may increase productivity
and lower costs while reducing hazardous wastes.
Much research and development work remains to be
done to bring many of these innovations to fruition.
A later report from this OTA study will examine the
extent to which improved technologies for energy
production and use can contribute to economic
development and environmental protection; under
what circumstances energy technology innovations
might ease the difficult tradeoffs between economic
growth and the environment; and policy issues
associated with accelerating the introduction of
improved energy technologies that minimize envi-
ronmental degradation.
This chapter traces the causes and consequences
of environmental degradation in developing coun-
tries, beginning in rural areas and following them
through to urban areas, modern industry, and the use
of fossil fuels. This organization has several advan-
tages over the more conventional cataloging of
environmental problems by their impacts on land,
water, and air. First, it helps capture the dynamic
causes of environmental degradation in developing
4 U.S. Congress, Office of Technology Assessment Enhancing Agriculture in Africa: A Role for U.S. Development Assistance, OTA-F-356
(Washington, DC: U.S. Government Printing mice, September 1988). Some note, however, that even steep or acid-infertile lands can be productive
over long periods as shown by the centuries of terraced rice farming in Asia or continuous sugar-cane cropping in the Dominican Republic. ScRicardo
Radulovich, "A View on Tropical Deforestation, "Nature, vol. 346, July 19, 1990, p. 214.
Chapter 5-Energy and the Environment in Developing Countries • 113
countries in the various contexts of the rural struggle
for survival by the landless peasant, or of the urban
effort to develop modern industry in an economy
that has limited capital, technology, and skilled
human resources. Second, it highlights the differ-
ences in the causes and consequences of environ-
mental degradation between rural and urban areas.
Third, it helps illustrate some of the complex
linkages between these environmental problems that
make effective responses so difficult. If economic
needs are to be met while simultaneously maintain-
ing environmental quality, these dynamics and
complex interconnections must be understood and
responded to appropriately.
The Rural Sector
Many developing countries are suffering signifi-
cant environmental degradation in rural areas, in-
cluding: rapid devegetat ion/deforestation and the
loss of irreplaceable biodiversity; desertification;
erosion of crop and pastureland; watershed degrada-
tion, siltation of downstream waterways, down-
stream flooding, and other impacts on water quality
and supplies; and local and regional air pollution. In
turn, these environmental impacts may damage or
destroy farm and range land and force more people
into watersheds and other ecologically fragile lands-
potentially creating a vicious cycle of degradation.
Clearing and burning woodlands and grasslands for
agriculture or pasture also contribute to the global
increase in atmospheric greenhouse gases.
The principal direct causes of these forms of
environmental degradation are shifting cultivation,
conversion of forest lands to permanent pasture and
agriculture, commercial logging, overgrazing, inap-
propriate management of irrigated land, and the use
of biomass for fuel. These factors are often intercon-
nected. Commercial logging of closed tropical
forests, for example, opens new areas with roads and
partial clearings, which enables settlers to follow,
converting forests to pasture and cropland.
Underlying these causes are factors such as:
population growth; poverty; the lack of access to
modern energy-intensive inputs for agriculture (me-
chanical traction, irrigation, fertilizers, etc.); the lack
of secure land tenure for many people; and govern-
ment pricing, tax, and other policies that may
encourage misuse of natural resources. These causes
and consequences are listed in table 5-1.
The extent and rate of environmental degradation
are difficult to quantify accurately under the best of
circumstances, even when reliable data are avail-
able. As it is, estimates vary widely on the basis of
the underlying definitions of what constitutes eco-
logical vulnerability or degradation; the methodolo-
gies used for the analysis; and the assumptions used
to go from local measures of degradation to global
extrapolations.
The depiction of environmental degradation also
depends on what measure is used for comparison.
Local degradation may be severe and rapid. Irrespec-
tive of the hardship this places on people in the
immediate region and the need for local responses,
it does not necessarily translate into corresponding
problems at the regional or global scale. There is
considerable variation in form and degree of envi-
ronmental degradation from one region to another.
Consequently, an attempt is made below to place
estimates of the rate of local environmental degrada-
tion into the global context.
Causes of Environmental Degradation
Agriculture
As the populations of developing countries grow,
the demands on the land for food, fuel, and fodder
increase accordingly. Farmers then face three basic
choices: they can 'mine' the land— taking more out
of it than they put in-until the land is exhausted;
they can migrate to new lands; or they can increase
the level of (capital-, energy-, and labor-intensive)
agricultural inputs-mechanical traction, fertilizer,
and irrigation— into the land in order to raise yields.
These might also include higher inputs of informa-
tion and management as might be the case for
intercropping, agroforestry, integrated crop-live-
stock, or other sophisticated agricultural systems. 5
Mining the Land — "Mining" the land usually
takes the form of shortened fallow periods-leaving
inadequate time for the natural regeneration of
soil-as population pressures mount. It is obviously
a temporary solution, but one often resorted to by
those without access to the modern agricultural
5 U.S. Congress, Office of Technology Assessment, Enhancing Agriculture in Africa: A Role for U.S. Development Assistance, OTA-F-356
(Washington, DC: Government Printing Office, September 1988).
114 •Energy in Developing Countries
inputs needed to raise crop yields. 6 Declining yields
from such "mining" are seen in a number of local
and regional areas, particularly in Africa. 7
Migration-Many people migrate to new lands or
to urban areas. In addition to population pressures or
soil exhaustion, factors forcing people to migrate
from long-established farming areas to new lands
include the low productivity of traditional agricul-
ture; inequities in land tenure for many subsistence
farmers; drought or other disasters; and, in some
areas, mechanization of agriculture and/or consoli-
dation of agricultural lands. In many cases, currently
farmed areas have been subdivided among succes-
sive generations to the point that the landholdings
for those who remain are, or will soon become,
marginal. In Rwanda, for example, the average
smallholder had 1.2 hectares in 1984 and, by
tradition, would divide it equally among his average
of four sons— leaving them 0.3 hectares each. If the
same trend continues, the following generation will
have less than 0.1 hectare each. 8 Estimates of the
number of landless or near-landless (with too little
land to subsist) in developing countries range as high
as 1 billion people, most of them in Asia. 9
Wage-paying jobs are scarce for those who are
forced to migrate. Land-anywhere they can get
it— for subsistence agriculture and fuelwood is often
their only means of survival. Increasingly, however,
available lands are remote, only marginally produc-
tive, or ecologically fragile— on upland regions that
are easily eroded when groundcover is removed, on
arid or semi-arid lands, or in forested areas of high
biological diversity (but which may have poor soils).
As many as 370 million people in developing
countries may live in rural areas that are ecologically
vulnerable. 10 To generate good yields on a sustain-
able basis on these lands often requires larger inputs
of labor and/or capital and technology than the lands
left behind-inputs to which these people seldom
have access.'This maybe particularly true in newly
opened areas where infrastructure (including access
to extension efforts) is especially weak. The immi-
grants into these areas may be unfamiliar with the
different agricultural techniques appropriate (sus-
tainable) to these new lands and resources.
In many regions, shifting agriculture is initially
practiced by those who migrate.^ (Sing agricul-
ture is also practiced traditionally by "long-term"
residents in many areas and is a sustainable form of
agriculture if fallow periods are sufficiently long.)
Shifting agriculture begins with forest-fallow sys-
tems in which small patches of land are cleared and
cultivated for a few years and then left fallow for as
long as two to three decades (see ch. 3). This remains
an important form of agriculture in west Africa,
southeast Asian hill communities, parts of South
America, 13 and elsewhere. "This form of shifting —
or slash and burn-agriculture is believed by many
to be the most important cause of secondary 15 forest
destruction and to be roughly comparable to com-
mercial logging in its impact on primary forest (table
5-2). 16 Shifting cultivation consumes enormous
amounts of biomass energy in the process of clearing
the forest. 17
Modern Agriculture-Those farmers with access
to good soils and water resources and modern
agricultural inputs can increase the yields of their
croplands. Modern agricultural practices in develop-
6 Poor soil quality or inadequate water resources may also be significant constraints on raising crop yields even with high levels Of agricultural inputs.
''U.S. Congress, Office of Technology Assessment op. cit., footnote 5, pp. 63 ff.; United Nations Food and Agriculture Organization, African
Agriculture: The Next 25 Years, Annex n, "The Land Resource Base" (Rome, Italy: United Nations, 1986).
'United Nations Population Fund, "The State of World Population 1990," New York, 1990.
'H.Jeffrey Leonard, Environment and the Poor: Development Strategies for a Common Agenda (New Brunswick, NJ: Transaction Books, 1989).
I0 Ibid.
n U.S. Congress, Office of Technology Assessment, op. cit., footnote 5.
^Alternatively those who migrate may continue their previous pattern of agriculture — often permanent agriculture-rather than adopting shifting
agriculture techniques as traditional in the new area.
13 Norman Myers, Deforestation Rates i„ Tropical Forests and Their Climatic Implications (London: Friends of the Earth Limited, 1989).
14 U.S. Congress, Office of Technology Assessment, Changing By Degrees: Steps To Reduce Greenhouse Gases, forthcoming.
15 Secondary f . st is that which has been logged in the past and then allowed to regrow, or has otherwise been significantly affected by human activity.
16 See also Julia C. Allen and Douglas F. Barnes, "The Causes of Deforestation in Developing Countries," Annals of the Association of American
Geographers, vol. 75, No. 2, 1985, pp. 163-184.
17 KirkR. Smith, "The Biofuel Transition," Pacific and Asian Journal of Energy, 1987, PP . 13-32; Terry Rambo, "Human Ecology Research on
Tropical Agroecosystems in Southeast Asia," Singapore Journal of Tropical Geography, vol. 3, No. 1, 1982. Some of the biomass ash generated serves
as nutrients for the crops subsequently planted.
Table 5-2 — Estimates of Land Use Changes in Closed and Open Tropical Forests, circa 1980 (million hectares per year)
Closed forest Open forest
Seller and Food and Seller and Food gnd
Crutzen Agricultural Crutzen Agricultural
Land use change Low High Organization Lanly Myers Low High Organization Lanly
Primary forest to:
Shifting cultivation 2.6 3.6 3.2 3.4 1.9 — — 1.2 1.7
Permanent pasture ., „ 1.6 1.4 1.1 2.5 0.6 1.7 2.1 0.8 1.3
Permanent agriculture 0.3 2.2 1.1 2.3 0.0 0.2 3.0 0.7 0.8
Logged forest — - 4.6 3.7 4.5 — — — —
Logged forest to:
Permanent pasture — — 1.1 — 0.6 — — — —
Permanent agriculture — — 1.0 — 3.9 — — — —
Secondary forest to: 14.9 40.0 18.5 22.0 3.4 6.9 21.9 11.4 18.6
Shifting cultivation
Permanent pasture ..., 0.5 1.5 0.1 — 3.3 1.0 1.0 — —
Permanent agriculture 0.6 0.8 0.1 — 6.8 0.2 1.4 — —
NOTE: The data in this table maybe significantly inaccurate. The table is presented here only to provide a general indication of the rates and causes of deforestation. The Food and Agricultural
Organization (FAO) and Lanly data are essentially the same (Lanly is the principal author of the FAO study), based primarily on official government statistics, and maybe underestimates
of the rate of deforestation. A more recent review by Myers, for example, indicates substantially higher ratesThe UNFAO currently has underway a more detailed study that may resolve
some of these large discrepancies.
I Norman m DeforestathnRatesinTropicai Forests and Their Climatic Implications (London: Friends of the Earth, December 1989).
SOURCE: R.P.Detwilerand Charles A.S. Hall, "Tropical Forests and the Global Carbon Cycle," Science, vol. 239, Jan. 1, 1988, pp. 42-47, citing Seilerand Crutzen, Food and Agricultural
Organization, Lanly, and Myers.
I'
3
116 . Energy in Developing Countries
ing countries use, to varying degrees, improved
plant species, synthetic fertilizers, pesticides and
herbicides, irrigation, and mechanized operations to
generate higher crop yields. These require high
levels of capital and energy inputs.
Modern agriculture has, with mixed success, pro-
vided environmental, social, and economic benefits.
It has moderated cropland expansion into ecolog-
ically fragile or particularly valuable lands through
technological advance (the green revolution) and
energy inputs in the form of fertilizer, irrigation, and
mechanical operations. Over the 20-year period
from 1965 to 1985, cropped areas increased by only
14 percent, 35 percent, and 4 percent in Africa, South
America, and Asia while their populations increased
by roughly 75 percent, 60 percent, and 50 percent
respectively (table 5-3). Modern agriculture has also
raised the personal incomes of many farmers; and it
has contributed to national economic growth in
many countries, especially in Asia.
Modern agriculture has also had serious short-
comings. It has increased economic inequities be-
tween those farmers who have sufficient land and
access to capital and other inputs necessary for
high-yield agriculture and those farmers who do not
have such resources, and it has displaced laborers in
many cases. It has caused environmental damage
through the misuse of fertilizers, pesticides, and
herbicides. Inadequate investment and poor man-
agement have led to waterlogging and salinization of
valuable irrigated lands. Finally, modern agricul-
tural techniques require dramatic increases in com-
mercial energy use. 18 Concern over environmental
impacts and high dependence on purchased inputs
has led to considerable interest in farming systems
that depend more on resources internal to the
farming system and less on external purchases.
These agricultural and agroforestry systems tend to
be very information and management-intensive. 19
Irrigated Lands— Irrigation is an important ele-
ment in modern agriculture. It frees the farmer from
dependence on irregular rains and raises yields,
allowing double- and even triple-cropping. Some
160 million hectares of land in developing countries
are irrigated. In Asia alone, 100 million hectares are
irrigated, and this land produces roughly 60 percent
of the region's food on just 45 percent of its cropped
area. 20 ln India, more than 6 million electric and 3
million diesel pump sets have been deployed (see ch.
3 ) , consuming nearly 2,000 GWh of electricity and
3 million tons of diesel fuel in 1985. 21
Inadequate investment and poor management
have resulted in various degrees of salinization
and/or waterlogging of irrigated lands in many
countries. For example, by one estimate 75 percent
of Pakistan's irrigated land suffers salinization
and/or waterlogging, with corresponding reductions
in crop yields. 22 Some 20 million hectares-rough I y
half— of India's irrigated croplands have sufficient
salt buildup to reduce productivity; another 7 million
hectares of land in I ndia now lie unused due to
excessive salt. 23 (Similar problems afflict the United
States, where 20 to 25 percent of the 20 million
hectares of irrigated lands are affected by saliniza-
tion. 24 ) Reclamation is possible through improve-
ments in canals and other infrastructure to reduce
leakage and by providing drainage from the fields,
but it is expensive.
The technology to prevent or minimize saliniza-
tion and waterlogging has been available since the
18 Note that the energy intensity of shifting — or slash and burn-cultivation may be significantly higher than commercialagriculture when the energy
consumed by burning off the standing biomass is taken into account. However, commercial fuels are little used, if at all, iishifting agriculture.
•'Advisory Committee on the Sahel, Agroforestry in the West African Sahel, Board on Science and Technology for International Development
National Research Council (Washington DC: National Academy Press, 1983, 1984); U.S. Congress, Office of Technology Assessment, op. cit., footnote
5; Clive A. Edwards et al., Sustainable Agricultural Systems (Ankeny, Iowa: Soil and Water Conservation Society, 1990) ; Robert Winterbottom and
Peter T. Hazlewood, "Agroforestry and Sustainable Development: Making The Connection," AMBIO, vol. 16, No. 2-3, 1987, pp. 1(W110; C. Okail
and J.E. Sumberg, "Sheep and Goats, Men and Women: Household Relations and Small RuminaiProduction in Southwest Nigeria," Understanding
Africa's Rural Households and Farming Systems, Joyce Lewinger Moock (cd.) (Boulder, CO: Westview Press, 1986).
20 Montague Yudelman, "Sustainable and Equitable Development in Irrigated Environments," Environment and the Poor: Development Strategies
for a Common Agenda, Jeffrey Leonard (cd.) (New Brunswick, NJ: Transaction Books, 1989).
21 Ashok Desai, ' 'Energy Balances for India, 1985-86," contractor report prepared for the Office of Technology Assessment, 1990. This is equivalent
to 125,000 GJ.
22 Yudehnan, op. cit., footnote 20. Total cmp areas from World Resources Institute, World Resources 1990-91 (New York, NY: Oxford University
Press, 1990). For other estimates, see C.E. Whitman et al. (eds.), Rainfed Agriculture in the Near East Region, proceedings of the workshop at Amman
Jordan, Jan. 18-23, 1986, USDA and USAID, p. 92.
^Yudelman, op. cit., footnote 20. Total crop areas from World Resources 1990 -91, ' op. C it., footnote 22.
M World Resources Institute, World Resources 1987 (New York, NY: Basic Books, 1987), p. 280.
Table 5-3-Changes in Land Area Uses
Cropland area Permanent pasture Forest/woodlands Other land
1964-66 to 1964-66 to 1964-66 to 1964-66 to
Total Population 1985-87 1983-85 1985-87 1983-85 1985-87 1983-85 1985-87 1983-85
area density, 1989 percent percent percent percent percent percent percent percent
(million (people per of total change of total change of total change of total change
Region hectares) thousand ha) land area in cropland land area in pasture land area in forest land area in other uses
Africa 2,965 212 6.2 14.0 26.7 -0.5 23.2 -8.6 43.9 3.6
N.America 2,139 197 12.8 7.7 17.2 -2.4 32.0 -5.9 38.0 1.5
S.America 1,753 166 8.0 35.2 27.0 9.4 51.6 -7.3 13.4 -1.4
Asia 2,679 1,139 16.8 4.2 25.3 -3.0 20.2 -4.6 37.7 -0.7
Europe 473 1,050 29^6 -5^0 T7J5 -43 33^2 7J 104 3.9
SOURCE: World Resources Institute, World Resources 1988-89 (New York, NY: Basic Books, 19SS) table 16.1 ; World Resources institute, Work/ Resources 1990-91 (New York, NY: Oxford
University Press, 1990), table 17.1.
f
I
5
3-
I
■3
2
118 . Energy in Developing Countries
1950's, yet designers frequently fail to incorporate
these improvements in their projects. Many factors
contribute to this failure. One factor may be that
designers tend to be overly optimistic in their initial
appraisals: they assume that waterlogging and sal-
inization will not occur for 20 to 30 years, and that
drainage systems and other capital-intensive im-
provements will not be necessary before then. This
lowers the apparent costs of their projects and may
improve the chances that the projects will be
approved. 25
Pasture-The opening up of lands for pasture is
a significant cause of deforestation (table 5-2)
particularly in Central and South America where
grazing lands have increased by more than 9 percent
during the past 20 years (table 5-3).
Overgrazing may also play a significant role in the
desertification of semi -arid regions. 26 Overgrazing
and trampling can quickly destroy the grass layer.
Without the protection of ground cover, topsoil can
be washed or blown away, lowering fertility. In
some areas, the full force of rain on the soil can bring
clay particles to the surface and cause surface
hardening and sealing that seeds cannot penetrate. 27
The end result of such processes can be desertifica-
tion. 28
Herders burn grasslands to encourage new
growth; numerous studies have shown this new
growth to be particularly good forage for their
herds. 29 More generally, repeated burning is be-
lieved to help create and maintain much of the
world's savannah and grassland. 30 Such brushfires
in the African grasslands may burn as much as 2
billion tons of biomass annually, cause volatilization
of organic nitrogen and sulfur, and allow excessive
leaching of other valuable nutrients. 31 This maybe
particularly damaging in much of the Sahel, where
growth is already strongly limited by the lack of
nutrients. 32 Thus, brushfires help the herder feed his
animals in the near term but, in the longer term,
lower soil fertility, and kill brush and trees that hold
the soils and pull nutrients up from deep in the
ground. 33 Brush and grassland fires may also be
significant contributors to regional air pollution and
may contribute modestly to the global increase in
greenhouse gases. 34
More stable supplies of forage might reduce the
need of herders to maintain large numbers of animals
in order to ensure the survival of a few through
periods of drought. Higher quality forage (higher
protein content) would reduce the need to burn
grasslands. Inputs of capital- and energy-intensive
fertilizer, increased supplies of water, and mechani-
cal harvesting of the forage (or even grain crops)
when its protein content is at a maxi mum-as is
common in industrial countries-might aid in achiev-
ing both of these goals.
Commercial Logging-Commercial logging im-
pacts perhaps 3 to 5 million hectares of primary
tropical forest annually (table 5-2). In many areas,
only the highest grade logs are removed from the
forest. But for every tree removed, roughly 5 to 10
other trees are destroyed. 35 Commercial logging also
develops roads that allow settlers access to forested
regions, where they can clear the forests for farms or
ranches.
Biomass Fuels— Biomass-wood, crop residues,
and animal dung-is the primary fuel for people in
"Yudelman, «P- cit., footnote 20.
^William H. Schlesinger et al, "Biological Feedbacks in Global Desertification," Science, vol. 247, Mar. 2, 1990, pp. 1043-1048.
^National Research Council Board On science andTecbiiologyinDevelopment,A#roForej«ryinrteW'M/A^can5aAe/(WasbJngton,DC: National
Academy of Sciences, 1983); Georges Novikoff and Mohamed Skouri, "Balancing Development and Conservation in Pre-Saharan Tunisia," AMBIO,
vol. 10, Nos. 2-3, 1981, pp. 135-141. A f *
^United Nations Food and Agricultural Organization, ./\1I ICclll Agriculture: The Next 25 Years, Annex n, "The Land Resource Base," Op. Cit.,
footnote 7.
^J.Dirck Stryker, "Technology, HumanPressure, and Ecology in the Arid and Semi- Arid Tropicsyin H. Jeffrey Leonard (cd.), op. cit., footnote 9.
soCarl Sagan, Owen B. Toon, and James B. Pollack, "Anthropogenic Albedo Changes and the Earth's Climate," Science, vol. 206, 1979, P P
1363-1368; Daniel Finn, "Land Use and Abuse in the East African Region," AMBIO, vol. 12, No. 6, 1983, pp. 296-301; DJ. Pratt and M.D. Gwynne
(eds.),Rangeland Management and Ecology in East Africa (Huntington NY: Robert E. Kreiger Publishing Co., 1977).
31 NationalResearchCouncil, Board on Science and Technology in Development, Environmental Change in the West African Sahel (Washington, DC:
National Academy of Sciences, 1983); World Resources Institute, World Resources 1990-91 (New York, NY: Oxford University Press, 1990).
32 H.Breman and c.T deWi ^ " Ran g e l ant ' Productivity ^ Exploitation in the Sahel," Science, vol. 221, 1983, pp. 1341-1347.
33 Stryker, op. cit., footnote 29.
^World Resources Institute, World Resources 1990-91 (New York, NY: Oxford University Press, 1990).
35 Freri Pearce, "Hit and Run mSarawak,"Afew Scientist, May 12, I990,pp. 46-49.
Chapter 5™Energy and the Environment in Developing Countries • 119
rural areas, and in many regions it is also important
for the urban poor (see ch. 3). Until recently, it was
widely believed that many developing countries
were on the verge of being transformed into treeless
deserts or mountain wastelands due to the demand
for fuelwood. These predictions were largely based
on the "gap" model, which assumed a freed
per-capita demand for fuelwood and a growing
population while at the same time the forest base was
declining due to agricultural expansion and over-
cutting of the forest. 36
In more recent studies, however, it has been found
that the use of fuelwood is highly elastic according
to its availability and the labor required to collect it
or, if traded, its price. When fuelwood is difficult to
obtain by foraging, people quickly resort to lower
quality wood, agricultural residues, or dung. More
generally, rural subsistence farmers cause relatively
little damage to forests, as they usually take only
deadwood or small limbs. They do not have the tools
to cut down large trees. Much of the wood they
collect is from hedgerows or other sources near their
farms. I n Kenya, for example, trees outside the forest
supply half the fuelwood demand. 37 And in West
J ava, one study found that three-fourths of all the
fuel collected came from within family courtyards
and gardens, and two-thirds of this fuel was branches
and twigs .38
In contrast to rural foragers, commercialized
fuelwood and charcoal operations tosupply urban
households, commercial facilities and industrial
operations often cut whole trees and can damage or
destroy forested areas. The impact of commercial
demands for fuelwood are limited, however, as users
will switch to fossil fuels when fuelwood becomes
scarce and prices climb (ch. 3). 39 The extent of the
damage to the forest resulting from commercialized
fuel demand will then depend on the distance
between the user and the forest, the size of the forest,
the size of the demand, the rate of regrowth, and
other factors. The use of biomass for fuel is not
usually a principal cause of deforestation. It does,
however, add additional pressure on forest re-
sources. In arid or semi-arid regions, where forest
growth is slow and there is a high population density
or a concentrated urban demand for fuelwood, such
as the African Sahel, the use of biomass fuels can
contribute significantly to local deforestation. 40
Similarly, the use of biomass fuels can lead to local
deforestation in some cases where there is a concen-
trated commercial or industrial demand.
Environmental Impacts
The environmental impacts of agriculture, ranch-
ing, lumbering, and the use of biomass for fuel
include deforestation, desertification, soil erosion,
flooding, pesticide and fertilizer runoff from crop-
land, and air pollution from biomass fuels.
Deforestation 41
The forests of the developing world provide a
number of resources and benefits, including food,
fuel, fodder, fiber, timber, and medicines. Forests are
a vital ecological resource, protecting soils, main-
taining diverse plant and animal life, regulating the
flow of water, and playing an important role in the
global carbon cycle. Forests may also strongly
influence regional climates. 42 The loss of tropical
forests not only threatens to deprive the world of
valuable resources on which it currently depends,
but also to foreclose opportunities to discover new
^Gerald Foley, "DiscussionPaper on Demand Management," proceedings of the ESMAPEasternandSouthemAfricaHousehoW Ener gy Planning
Seminar, Harare Zimbabwe, Feb. 1-5, 1988, UNDP/World Bank Energy Sector Management Assistance Program, Activity Completion Report No.
085/88.
37phii O'Keefe, "Fuel for the People: Fuelwood in the Third World," AMBIO, vol. 12, 1983, pp. 21-26.
38 M. Hadi Soesastro, ' 'Policy Analysis of Rural Household Energy Needs in West Java, ' ' Rural Energy to Meet Development Needs: Asian Village
Approaches, M. Nurul Islam, Richard Morse, and M. Hadi Soesastro (eds.) (Boulder, CO: WestView Press, 1984), p. 114.
^Douglas F. Barnes, < 'Understanding Fuelwood Prices in Developing Nations," draft, Household Energy Unit, Industry and Energy Department,
World Bank, Washington DC, Oct. 31, 1989.
^Douglas F. Barnes, World Bank, "Population Growth, Wood Fuels, and Resource problems in Sub-Saharan Africa," Industry and Energy
Department Working Paper No. 26, March 1990; R. Moss and W. Morgan, Fuelwood and Rural Energy Production and Supply in the Humid Tropics
(Dublin, Ireland: Tycooly International Publishing Ltd., 1981); Finn, op. cit., footnote 30; Dennis Anderson and Robert Fishwick, World Bank,
"Fuelwood Consumption and Deforestation in African Countries," World Bank Staff Working Paper, No. 704, 1984.
41 For detailed reviews of deforestation and policy responses to it, see: U.S. Congress, Office of Technology Assessment, Changing By Degrees: Steps
To Reduce Greenhouse Gases, forthcoming; U.S. Congress, Office of Technology Assessment, Technologies to Sustain Tropical Forest Resources,
OTA-F-214 (Springfield, VA: National Technical Information Service, March 1984); and Robert Winterbottom, "Taking Stock: The Tropical Forestry
Action Plan After Five Years," World Resources Institute, Washington, DC, June 1990.
^J.Shukla, C. Nobre, and P. Sellers, "Amazon Deforestation and Climate Change," Science, vol. 247, Mar. 16, 1990, pp. 1322-1325.
120 . Energy in Developing Countries
potential sources of wealth and scientific knowl-
edge. On the other hand, tropical forests offer an
economic opportunity that the poor in developing
countries wish to seize as quickly as possible. 43
Governments may also want to colonize forest lands
in order to establish more clearly their legal claim to
national territory.
The rate of global deforestation -from all causes-
is highly uncertain, with estimated rates in the early
1980s ranging from roughly 0.5 percenf'to 1.8
percent 45 annually. These rates appear to be acceler-
ating due to increasing population and other pres-
sures noted above.
The UNFAO 1990 Forest Resources Assessment
estimates the current annual global deforestation
rate as 1.2 percent-double their estimate for
1980. * Forested areas and deforestation rates vary
widely between countries. Indonesia and Brazil have
huge areas of closed forests (100 and 350 million
hectares, respectively) and annual rates of deforesta-
tion of perhaps 0.4 to 1.4 percent and 0.5 to 2.3
percent, respectively. Aggregate figures, however,
tend to obscure severe deforestation occurring in
certain regions. Ivory Coast, for example, has just 16
million hectares of forest remaining, and suffers an
annual deforestation rate variously estimated be-
tween 6.5 and 15.6 percent. A number of other
countries lie between these extremes, with forested
areas of 5 to 50 million hectares and deforestation
rates variously estimated in the range of 2 to 8
percent annually .47 At these rates, their closed
forests could disappear in a few decades.
Desertification
Desertification can result from a variety of factors,
depending on the region, including long-term cli-
mate trends, overgrazing, poor farming practices,
and deforestation. 48 Although anecdotal evidence
indicates that drylands in many regions are becom-
ing decertified at an increasing rate, there is little
reliable data to support the case. The "Global
Assessment of Soil Degradation," initiated in late
1987 by the United Nations Environmental Program
and the International Soil Reference and Information
Centre in the Netherlands, should provide some of
these data.
Impacts of Deforestation and Desertification
Among the potential impacts of deforestation and
desertification are soil erosion and degradation,
fuel wood and fodder shortages, increased flooding,
microclimatic changes, and loss of biodiversity. 49
Soils— Little soil is lost from forests or grass-
lands. When vegetation is removed, massive amounts
of soil can be washed away as rainwater flows across
the surface. Measurements in Tanzania indicated
that up to half the rainfall was lost as runoff from
bare fallow (3.5° slope), carrying with it some 70
tons of soil per hectare. 50 Similar impacts have been
noted elsewhere. 51 With no shading, soil tempera-
tures rise dramatically and can greatly reduce the
vital biological activity in the soil. 52 Loss of tree
cover also allows higher average wind velocities
(and soil erosion) and, combined with the reduced
soil moisture content, can lower crop yields. 53
<3 RicardoRadulovich, "A View on Tropical Deforestation," Nature, vol. 346, July 19, 1990, p. 214.
^Jean-Paul Lanly, "Tropical Forest Resources, « Forestry Paper No . 30, United Nations Food and Agriculture Organization, Rome 1982.
^Myers, op. Cit., footnote 13-
" 'World Deforestation Increases at Quicker Rate Than Expected," MultinationalEiivironmental Outlook, Aug. 21, 1990, p. 134.
47 Myers, °P- cit -' footnote 13, P- '* 4 o
48q Novikoff , '~ Desertification by Overgrazing,' ' AMBIO, vol. 12, No. 2, 1983, pp. 102- 105; H.F. Lamprey and Hussein Yussuf, ' 'Pastoralism and
Desert Encroachment inNorthernKenya," AMBIO, vol. 10, Nos. 2-3,1981, pp. 131-134; NationalResrarcA Council, Environmental Change in the West
African Sahel, op. cit., footnote 31 ; D. Anderson, and R. Fishwick, "Fuelwood Consumption and Deforestation in African Countries," World Bank,
Staff Working Paper No. 704, 1984; Schlesinger et al., op. cit., footnote 26.
49 KuswataKartawinataetal.," TneIm P act o f Manof a Tro P lcalForestmIndonesia ." AWB/0 .vol.lO,Nos. 2-3,1981, PP . 115-119; Lester R. Brown,
"World Population Growth, Soil Erosion, and Food Security," Science, vol. 214, 1981, pp. 995-1002; Alain Grainger, Desertification (London:
Earthscan, 1984), p. 94.
sophil OTCeefe, "The Causes, Consequences and Remedies of Soil Erosion in Kenya," AMBIO, vol. 12, No. 6, 1983, pp. 302-305.
51 U.S. Congress, Office of Technology Assessment, op. cit, footnote 41; Finn, op. cit., footnote 30; Eneas Salati and Peter B. Vose, "Depletion of
Tropical Rainforests;" AMBIO, vol. 12, No. 2, 1983, pp. 67-71, Vaclav Smil, "Deforestation in China," AMBIO, vol. 12, No. 5, 1983, pp. 226-231;
Nigel JJH. Smith, "Colonization Lessons from a Tropical Forest," Science, vol. 214, 1981, pp. 755-761; A.H. Gentry and J. Lopez-Parodi,
"Deforestation and Increased Flooding of the Upper Amazon," Science, vol. 210, 1980, p. 1354.
52 Salati and Vose, op. cit, footnote 51; Henri Dosso, Jean Louis Guillaumet and Malcolm Hadley. "Land Use Problems in a Tropical Forest,"
AMBIO, vol. 10, No. 2-3, 1981.
^Dennis Anderson, "Declining Tree Stocks in African Countries," woWtf Development, vol. 14, No. 7, 1986, pp. 853-863.
Chapter 5™Energy and the Environment in Developing Countries . 121
Fuel and Fodder — As forests and grasslands
disappear, rural people are increasingly forced to
rely on agricultural residues and dung for their
cooking and heating needs, and on crop residues for
animal fodder. The failure to return organic materi-
als to the soil can have significant environmental
impacts, even if these impacts are longer term and
more subtle than is sometimes suggested. People in
many areas already divert organic residues to other
uses, often with little apparent near-term effect on
yields. For example, crop residues such as millet or
sorghum stalks tend to be poor fertilizers and are
difficult to recycle; they are often burned in the
fields to prevent them from harboring crop pests.
Similarly, dung quickly loses its nitrogen and much
of its effectiveness as a fertilizer when left lying in
the SUn, as is common. 54 In areas with poor soils
and/or high rainfalls that quickly leach nutrients out,
however, crop yields may drop quickly if residues
are not returned to the soil.
I n the longer term, the loss of organic material can
reduce the productivity of even the highest quality
soils. Organic matter in soils provides important
nutrients needed by plants; it helps the soil bind
important minerals-e. g., magnesium, calcium, and
potassium-that would otherwise be leached away;
it buffers the acidity of the soil; and it improves
water retention and other physical characteristics. 55
Wafer— when the natural water regulation sys-
tem provided by forests and grasslands is removed,
stream flows tend to become more erratic, with
reduced flows during dry seasons and worse floods
in the wet season. This can interfere with agriculture,
fishing, and darns and can threaten inhabitants.
Lifetime (years)
Reservoir
Planned
Revised
Bhakra
Maithon
Hirakund
Ram Ganga
88
246
100
185
47
24
35
48
Table 5-4-Sedimentation Rates of Some
Reservoirs in India
Lifetime
Planned
88
246
100
... 185
SOURCE: Kunwar Jalees, "Loss of Productive Soil in India," International
Journal of Environmental Studies, vol. 24,1985, pp. 245-250.
Eroded soils choke downstream waterways and
reservoirs, reducing their ability to handle the
increased volumes of water running directly off the
watersheds. 56 Over the past 10 years, the area
annually flooded in India has increased by 18
percent. Some observers attribute this increased
flooding to the clearing of regional forests. Flood
and erosion damage due to the clearing of I ndia's
forests has been estimated at $20 billion for the
period from roughly 1960 to 1980, including loss of
topsoil, loss of property to floods, and shortened
reservoir lifetimes (table 5-4). 58 0ther estimates
place the direct costs of repairing flood damage in
India at more than $250 million per year. 59
Climate — In some regions, a significant portion
of the rainfall is generated from moisture pumped
back into the atmosphere by vegetation. Removal of
this vegetation may then contribute to climatic
change in the region. 60 The surface reflectance is
changed and may likewise affect cl i mate. 61 To the
extent that the local climate changes due to the loss
of vegetation, it may become more difficult to
reverse the process and restore grasslands and
^G.C.Aggarwal, "Judicious Use of Dung in The Third World," Energy, vol. 14, No.6,1989,PP. 349-352.
^Geoffrey Barnard and Lars Kiistofeison, Agricultural Residues as Fuel in The Third World (London: Earthscan, 1985.
^Erik P- Eckholm, Losing Ground: Environmental Stress and World Food Prospects (New York, NY: W.W. Norton& Co., 1976).
57 U.S. Congress, Office of Technology Assessment, op. cit., footnote41.
M rbid.AmoregeneralreviewofuusprobleminIndiais given in The Stateoflndia'sEnvironment 1984-84; The SecondCitizen'sReport (New Delhi:
Center for Science and Environment).
59 John Spears, "Preserving Watershed Environments," t/2VASm», vol. 34, No. 137, 1982, pp. 10-14.
^Sagan et al., op. cit., footnote 30; J. Shukla and Y. Mintz. ' 'Influence of Land-Surface Evapotranspiration on the Earth's Climate," Science, vol.
215, 1982, pp. 1498-1501.
61 RobertS.Kandel, Mechanisms Governing the Climate of the Sahel: A Survey ofRecent Modelling and Observational Studies, OECD/CTLSS/Club
du Sahel, Report Sahel D(84)252, October 1984. See also Sagan et al., op. cit., footnote 30, estimates oialbedo changes are given in Gregory Morris,
"Environmental Impacts of Bioenergy Development," Biomass Energy Systems: Building Blocks for Sustainable Agriculture, Jan. 29-Feb. 1, 1985,
Airlie House, Airlie, VA, The Rockefeller Brothers Fund and World Resources Institute, Washington, DC.
122 . Energy in Developing Countries
forests to something approaching their original form
and extent. 62
Biodiversity 63 -Plants in tropical rain forests
have evolved a particularly rich array of potentially
useful chemicals, yet fewer than 1 percent of all
tropical plant species have so far been screened for
biochemical of use to man. 64 A number of valuable
medical products have already been developed from
tropical plants, including highly effective drugs for
leukemia and hypertension, muscle relaxants, and
others. 65 Many other aesthetic, scientific, and com-
mercial benefits may be realized from tropical forest
products. Tropical deforestation and the loss of
species would foreclose many of these opportuni-
ties. 66
Although they have received much less attention,
plants in arid and semi-arid regions similarly offer
many potential benefits. These areas harbor a
number of wild relatives to our basic crops-beans,
potatoes, wheats, barleys, millets, sorghums, and
many others— and are sources for genetic resistance
to drought, heat, pests, and disease. The diverse
genetic resources of wild varieties can be an
important input into the higher yielding but geneti-
cally narrow varieties used in intensive agriculture.
Many valuable genetic strains and species may beat
risk in these arid and semi-arid regions. 67
Deforestation can result in a direct conflict
between the survival of species and the livelihoods
of people who exploit the land. Such a conflict
recently erupted in the United States over the
endangered northern spotted owl. 68 There are similar
conflicts in many developing countries. Tropical
rain forests contain at least half and perhaps as many
as 90 percent or more of the species found on earth,
yet cover just 7 percent of the total land area. No one
knows precisely the share of species held by tropical
rain forests, because fewer than 2 million species
have been officially classified out of a total number
that is variously estimated to range from less than 5
to more than 50 million. 69 Many countries are taking
steps to protect their biodiversity, and a few are
realizing some success. Protection of the brow-
antlered deer in northeast India, for example, is
helping it make a comeback from near extinction. 70
There are concerns in developing countries,
however, that the costs of conserving biodiversity —
in jobs and/or access to land and resources for the
poor, security enforcement, administration, and the
mistrust generated as governments deny their own
people access to much-needed resources-will be
born by the developing countries, but the benefits
will be largely captured by commercial interests in
industrialized countries. 71
Air Pollution
The burning of biomass generates large amounts
of air pollution in developing countries. Food is
typically cooked over an open fire or a poorly vented
stove-exposing women and children, as well as
other members of the family, to high levels of toxic
smoke (ch. 3). Similarly, in colder climates, homes
in rural areas are often heated by open fires, further
increasing exposures to toxic smoke. 72 One impact
of this indoor air pollution is exacerbation of one of
the most deadly classes of infectious illness, acute
respiratory infections in children.
The use of biomass for fuel, clearing forest land
for agriculture, and burning grasslands to generate
Mj.Shukla, C. Nobre, P. Sellers, "AmazonDeforestation and Climate Change" Science, vol. 247, Mar. 16, 1990, pp. 1322-1325. H. Schlesinger et
al., op. cit., footnote 26.
^For a complete review of this complex issue, see U.S. Congress, Office of Technology Assessment, Technology to Maintain Biological Diversity,
OTA-F-330 (Springfield, VA: National Technical Information Service, March 1987).
"NeflA-Belson, ' 'Tropical Deforestation and the Response of the UnitetStates Congress," Georgetown International Environmental Law Review,
vol. 2, Fall 1989; U.S. Congress, Office of Technology Assessment, op. cit., footnote 41.
^U.S. Congress, Office of Technology Assessment, op. cit., footnote 41.
**Edward O. Wilson, "Threats l - Biodiversity," Scientific American, September 1989, PP . 108-116; Walter V.Reidand Kenton R. Miller, Keeping
Options Alive: The Scientific Basis for Conserving Biodiversity (Washington, DC: World Resources Institute, October 1989).
67 GaryNabhan, "How Are Tropical Deforestation and Desertification Affecting Plant Genetic Resources, "Annals ofEarth, vol. 4, No. 1, pp. 21-22,
1986; Paul Raeburn, "Seeds of Despair," Issues in Science and Technology, Winter 1989-90, pp. 71-76.
^library of Congress, Congressional Research Service, "Economic Impacts of Protecting the Northern Spotted Owl," 90-74 ENR, Mar. 5, 1990.
69 RobertM. May, "How Many Species Are There on Earth," Science, Sept. 16, 1988, pp. 1441-1449.
^Sanjoy Hazarika, "A Deer Comes Back From the Brink in India," New York Times, July 31, 1990.
71 Ricardo Radulovich, • 'a View on Tropical Deforestation," Nature, vol. 346, July 19, 1990, p. 214.
72 Kirk R - Smith, Biofuels, Ai , Pollution, and Health: A Global Review (New York, ny : Plenum Press, 1987).
Chapter 5~Energy and the Environment in Developing Countries • 123
fresh forage for livestock all generate large amounts
of smoke that contributes to regional air pollution.
These activities also pump greenhouse gases into the
atmosphere, potentially contributing to global cli-
mate chance.^ tms is discussed in much more OS-
tail in the forthcoming OTA report, Changing By
Degrees: Steps To Reduce Greenhouse Gases.
The Urban Sector
Causes of Environmental Degradation
Urban areas of developing countries are growing
rapidly, in large part due to migration from rural
areas. For the low-income countries, urban popula-
tion jumped from 17 percent of total population in
1965 to 30 percent in 1987. In 1960, there were 59
cities with more than 500,000 persons in developing
countries; by 1980 that number had grown to 165
cities. 74 ! n the early 1990's, cities in developing
countries will need to absorb more than 100 million
additional people annually. 75 This trend has impor-
tant implications for both energy use and energy-
related environmental impacts.
Urbanization and modernization can provide many
desirable benefits-improved standards of living,
increased opportunities for education and employ-
ment, and greater insulation from the vagaries of
drought endured by those in rural areas. (They also
reduce pressures on some aspects of the rural
environment.) They also have costs, including
potential damage to the environment.
Urbanization changes the consumption patterns
of goods and energy- related services (ch. 3). House-
holds make a transition from biomass to commercial
fuels for cooking and other domestic services.
Demand for consumer goods, notably electric appli-
ances, increases. The economic base changes from
agriculture and small rural industry to larger manu-
facturing and services, with a corresponding change
in the demand for commercial energy. The need for
public and personal transport grows as employment
shifts from agriculture or rural industry located
within walking distance of residences to large
industry and commerce located further away. The
high concentration of people requires the transport
of food, fuel, and other materials from long dis-
tances, as well as effective management of wastes.
These changes have significant impacts on energy
use and on the environment.
Residential/Commercial Sector
The transition from biomass to clean commercial
fuels such as liquefied petroleum gas (LPG) and
electricity for residential cooking generally reduces
the total amount of air pollution emitted and largely
shifts that which is emitted from the household to
distant refineries and electricity generation plants
(ch. 3). 76 l n some countries, however, clean com-
mercial fuels are not widely available or remain too
expensive. China, for example, uses about one-third
of its coal in residences, of which nearly half is used
for cooking (app. 3-A). Due in part to heavy
residential coal use, some northern Chinese cities,
such as Beijing and Tianjin, have very high sulfur
dioxide concentrations.
Lights, refrigerators, air conditioners, and other
electrical appliances in the residential/commercial
sector provide highly desirable services and are
penetrating urban areas much more rapidly than
rural areas (ch. 3). These appliances consume large
amounts of electricity, however, the generation of
which can cause significant environmental damage
if not properly controlled. Refrigerators and air
conditioners also use chlorofluorocarbons (CFCs),
which have already damaged the earth's protective
ozone layer and are potent greenhouse gases.
Industrial Sector
Industry provides many goods that contribute to
our material comfort and well-being. Industry can
damage the environment through a variety of
activities, however, if they are mismanaged or
inadequately controlled. These include: placing
heavy demands on natural resources as feedstocks or
other inputs; intensively using electricity for me-
chanical drive and other needs; burning fossil fuels
73 R.P. Detwiler and Charles A.S. Hall, "Tropical Forests and the Global Carbon Cycle," Science, vol. 239, Jan. 1,1988, PP- 42-47; Richard A.
Houghton, "TheFutureRoleof Tropical Forests in Affecting the Carbon Dioxide Concentration of the Atmosphere." AMBIO, vol. 19, No. 4, July 1990,
pp. 204-209.
74 World Bank, World Development Report 1989 (New York, NY: Oxford University Press, 1989).
1! WorldDevelopment Report 1990, p. cit.; Indicator Tables l and 3 1. The urban population — 41 percent of the developing country total of 3.95 billion
people-1.6 billion people and is growing at 6.9 percent annually.
76 Tbis refers to carbon monoxide particulates, sulfur dioxide, nitrogen oxides, and other such chemical species. It does not refer to carbon dioxide,
which may be increased by the use of nonbiomass fuels but not by biomass if the biomass is being produced on a sustainable basis.
124 . Energy in Developing Countries
for process heat; and generating hazardous wastes
that may be discharged into landfills, water systems,
or the air. Each industry has a different mix of these
activities, and each activity has a different set of
environmental impacts, depending on the use of
environmental controls and other means of mitiga-
tion.
Industrial pollution can pose a particularly severe
health hazard in developing countries when indus-
tries are established near existing residential areas,
or when residential areas buildup around them. This
has led to tragedies such as Bhopal, as well as serious
systemic pollution such as found in Cubatao, Brazil,
in the 1970's and early 1980's. 77 Often, the lowest
income people are most seriously affected. This can
lead to the situation in which the poor face all of their
traditional risks to health-infectious disease, hun-
ger and malnutrition, air pollution from biomass
fuels-and at the same time face even greater than
normal modem risks-such as exposure to hazard-
ous wastes and toxic air pollutants. 78
Transportation Sector
The transportation sector is the largest contributor
to air pollution in many cities. For example, in
Indian cities, gasoline-fueled vehicles— mostly two
and three wheelers— are responsible for 85 percent
of carbon monoxide and 35 to 65 percent of
hydrocarbons in the air from fossil fuels. 79 Diesel
vehicles— buses and trucks-are responsible for
over 90 percent of nitrogen oxide (NO x ) emissions
in urban lndia. 80 Use of emission control devices and
engine modifications have significantly reduced
exhaust emissions on newer vehicles in the United
States, but these pollution controls are usually not
standard on vehicles in developing countries. Runoff
from roads also contributes to water pollution.
In each of these sectors, there are often difficult
tradeoffs between longer term environmental im-
acts and immediate financial costs to consumers.
Many developing countries also have shortages of
skilled technical manpower to implement mitigation
efforts. The extent to which technological innova-
tions or other advances might ease these tradeoffs
will be examined in a later report of this OTA study.
Environmental Impacts
Air
Air quality in many of the developing world's
cities is poor, and has been deteriorating. The United
Nations Environment Program (UNEP) estimates
that up to one-half of the world's urban population,
including residents of many industrialized countries,
live in areas with marginal or unacceptable levels of
sulfur dioxide (S0 2 ) in the air. Concentrations of
S0 2 (see figure 5-1) and concentrations of total
suspended particulate (see figure 5-2) in major
cities in the developing world are considerably
above World Health Organization guidelines. Photo-
chemical smog has become a recurrent seasonal
problem in many large tropical and subtropical
cities.
The sources of these pollutants vary. Coal-used
for electricity generation, industrial power and
process heat, and domestic heating and cooking
(China)— primarily emits sulfur dioxide, particu-
late, and nitrogen oxides. The combustion of oil or
gas in stationary sources, such as electric generating
units, emits many of the same pollutants-nitrogen
oxides, hydrocarbons, particulate, and sulfur diox-
ide-but in much different proportions. 81 Nitrogen
oxide, a major ingredient of urban smog, is the most
harmful pollutant released on a large scale when
electricity is generated from oil or gas. Mobile
sources-cars, trucks, two and three wheelers, and
buses-release large amounts of carbon monoxide,
NO=, hydrocarbons, and particulate and are typi-
cally the largest source of these pollutants in urban
areas. These mobile sources are also major contribu-
tors of lead pollution due to the use of lead as an
octane booster in gasoline. Burning fossil fuels
unavoidably generates carbon dioxide, the most
important greenhouse gas. These pollutants (except
carbon dioxide (C0 2 )) can damage crops, forests,
and structures and can aggravate human health
problems.
77 World Resources Institute, World Resources 1990-91 (New York, NY: Oxford University Press, 1990), p. 41.
7811^6 more general case, there is a shift in the types a n.i that people are exposed to from the traditional to the modem. See Kirk R- Smith, "The
Risk Transition," International Environmental Affairs, vol. 2, No. 3, in press.
'''Tata Energy Research Institute, TERI Energy Data Directory and Yearbook 1988 (New Delhi, India: 1989), p. 250.
«°Ibid.
81 Mitre Corp., ' 'Health and Environmental Effects of Oil and Gas Technologies: Research Needs," report to the Federal Interagency Committee on
the Health and Environmental Effects of Energy Technologies, Juljl 981 .
Chapter 5™Energy and the Environment in Developing Countries • 125
Figure 5-1-Sulfur Dioxide Levels in Selected Cities, 1980-84
Shown is the range of annual values at individual sites and the composite 5-year average for the city,
< Concentration (ug/m 3 )
Key
I.Milan
2.. Shenyang 3.Tehran
4. Seoul
5. Rio de Janeiro
6. Sao Paulo
7.Xian
8. Paris
9. Beijing
10. Madrid
11. Manila
12. Guangzhou
13. Glasgow
14. Frankfurt
15. Zagreb
16. Santiago
17. Brussels
18. Calcutta
19. London
20. New York City
21. Shanghai
22.Hong Kong 23. Dublin
24. St. Louis
25. Medellin
26. Montreal
27. New Delhi
28. Warsaw
29. Athens
30. Wroclaw
31. Tokyo
32. Caracas
33. Osaka
34. Hamilton
35. Amsterdam
36. Copenhagen
37. Bombay
38. Christchurch
39. Sydney
40. Lisbon
41. Helsinki
42. Munich
43. Kuala Lumpur
44. Houston
45. Chicago
46. Bangkok
47. Toronto
48. Vancouver
49. Bucharest
50. Tel Aviv
51. Cali
52. Auckland
53. Melbourne
54. Craiova
10
WHO Guideline 40-60 ug/m 3
/ \ 100
I
I
I
"I — I I I I I
Range of individual site
annual averages
Combined site
average 1980-84
T
"i — i — r
i i ii
L
J 3.
14.
I 5.
r;:
no.
111.
TT
1
|14.
1
f
1
[16.
I 17
I [20.
| I
l I
•
23.
24.
25.
I
I
I
126.
]28.
]29.
"30.
"31.
35.
' 36.
133.
P4
T39.
40.
r«i.
„ irfri.
ITT*
-42.
43.
47.
"150.
hk
a Note logarithmic scale.
SOURCE: World Health Organization and United Nations Environment Programme, Glo&IPollution and Health (London: Yale University Press, 1987),figure
2, p. 5. (Logarithmic scale.)
726 • Energy in Developing Countries
Figure 5-2-Suspended Particulate Matter Levels in Selected Cities, 1980-84
Shown is the range of annual values at individual sites and the composite 5-year average for the city.
10
r
WHO Guideline 60-90 ug/m 3
~ 1
1 — I I I II
T
t — r~rr-
Range of individual site
annual averages
i 1
I
L
J
Combined site
average 1980.84
I I [ 3-
I I 16.
I 9.
10.
111.
I 12.
I L
J 15.
1 I N
16.
/
t9.
20.
Key
1. Kuwait
2. Shenyang
3. Xian
4. New Delhi
5. Beijing
6. Calcutta
7. Tehran
8. Jakarta
9. Shanghai
10. Guangzhou
11. Illigan City
12. Bangkok
13. Bombay
14. Kuala Lumpur
15. Zagreb
16. Rio de Janeiro
17. Bucharest
18. Accra
19. Lisbon
20. Manila
21. Chicago
22. Caracas
23. Birmingham
24. Helsinki
25. Hamilton
26. Sydney
27. Houston
28. Craiova
29. Toronto
30. Melbourne
31. Medellin
32. Chattanooga
33. Fairfield
34. Montreal
35. Vancouver
36. New York City
37. Tokyo
38. Osaka
39. Cali
40. Copenhagen _! — ! 14 °-
41. Frankfurt 1 1 i 41 -
a Note logarithmic scale.
SOURCE: World Health Organization and United Nations Environment Programme, Global Pollution and Health (London; Yale University Press, 1987), figure
3, p. 6. (Logarithmic scale.)
Efforts to reduce emissions from coal burning
usually involve removing harmful material from
exhaust gases after combustion. Pollution controls,
such as electrostatic precipitators or baghouses to
control particulate, capture pollutants from exhaust
gases, but they create solid wastes in the process.
Flue gas desulfurization (FGD) equipment ("scrub-
bers") controls sulfur emissions by capturing the
sulfur in the form of liquid sludges. Electrostatic
precipitators are widely used in the developing
world, but FGD systems are not common. 82 This
may be due in part to the high costs of such systems.
One study in India found that the addition of an FGD
system would add about 15 percent to the cost of
82 Interiiational Development and Energy Associates, Inc., "Clean Coal Technologies for Developing Countries," contractor report prepared for the
Mice of Technology Assessment, May 1990, p. 25.
Chapter 5~Energy and the Environment in Developing Countries ^127
electricity .83 Perhaps as significant is that Indian and
Chinese coals have relatively low sulfur contents-
typical ly 0.5 to 2.0 percent compared to 3 to 6
percent for Eastern European and some U.S. Coals. 84
Adoption of more effective pollution control
technologies already common in the industrialized
world and improvements in energy efficiencies
might significantly improve air quality in many
cities of the developing world.
Water
The primary urban water pollution problem is the
discharge of untreated or minimally treated human
waste into water systems (this is also a significant
problem in rural areas). As of the mid-1980's, only
about half the developing world's urban population
had adequate sewage facilities, and perhaps three-
fourths had access to safedrinking water. This
situation contributes to the growth and spread of
pathogenic viruses, bacteria, and protozoa and is
directly or indirectly responsible for perhaps three-
fourths of all illness and 80 percent of all child
deaths. 85
Industrial wastes also contribute significantly to
water pollution in some areas. These range from
agricultural industry wastes to heavy metals (mer-
cury, cadmium, lead, and chromium) and chlorin-
ated hydrocarbons-to name only a few. These
pollutants can have serious impacts on water re-
sources all the way downstream to the ocean. Many
coastal areas are also being affected. 86 A recent
study found that World Health Organization (WHO)
standards for heavy metals-cadmium, lead, and
mercury-were exceeded in one-third to three-
fourths of the fish and shellfish samples taken from
Jakarta Bay, Indonesia. 87 Mitigation strategies that
might be applied include treatment of industrial
effluents and reduction of the generation of hazard-
ous materials themselves through improvements in
manufacturing processes. 88
Land
The major causes of soil pollution in urban areas
are the disposal of solid wastes and the deposition of
air pollutants such as S0 2 and particulate. The
focus here is on solid wastes. Low-income countries
generate much less urban refuse (around 0.5 kg per
capita per day) than industrial countries (0.7 to 1.8
kg per capita per day). 89 In large cities, however,
even modest per-capita levels can lead to large
amounts of waste generation. Mexico City is esti-
mated to generate 11,000 tons a day. Most of this
waste originates from households, largely in the
form of vegetable and putrescible materials. There is
also a significant component of hazardous wastes
(containing chemicals, heavy metals, and/or radio-
active wastes) from industries.
In the developing countries, current procedures
for disposing of such wastes are inadequate. Much
household garbage is not collected at all, but left to
decompose. Hazardous wastes may be dumped on
uncontrolled municipal landfills. Only a small part
of total wastes are disposed of in an environmentally
safe reamer. Nonetheless, costs of even inadequate
disposal can amount to as much as 30 percent of
municipal budgets. 90
Energy could, however, play a part in solving
these problems. Urban waste can be used as feed-
stock for cogeneration plants. Process changes in
energy-intensive industries such as steel, cement,
and fertilizer may well, by economizing on the use
of all materials, also reduce hazardous wastes.
"Ibid., p. 25.
M Ibid.,pp.7,8,27.
85 World Resources Institute, World Resources 1990-91 (New York, NY: Oxford University Press, 1990); World Resources Institute, "Development
and Pollution in the Third World," Tokyo Conference on the Global Environment and Human Response, Sept. 11-14, 1989; World Health Organization
The International DrinkingWater Supply and Sanitation Decade «s«n«v»: United Nations, 1986); World Health Organization and United Nations
Environment Programme, Global Pollution and Health (New Haven, CT: Yale University Press, 1987).
^United Nations Environment Programme, "The State of the Marine Environment," 1990.
87 World Resources Institute, "Development and Pollution in the Third World," op. cat., footnote 85.
M U.S. Congress, Office of Technology Assessment, Serious Reduction of Hazardous Waste For Pollution Prevention and Industrial Efficiency,
0TA-ITE-317 (Washington, DC: U.S. Government PrintingOffice, September 1 986). U.S. Congress, Office of Technology Assessment, From Pollution
to Prevention: A Progress Report on Waste Reduction, OTA-ITE-347 (Washington, DC: U.S. Government Printing Office, June 1987).
"'Sandra J. Cointreau, Environmental Management of Urban Solid Wastes in DevelopingCountries:AProjectGuide,qootedm ii De\elopment and
Pollution in the Third World," prepared by The World Resources Institute for the Tokyo Conference on the Global Environment and Human Response,
Sept. 11-14, 1989 (Washington, DC: World Resources Institute, July 1989).
"World Resources Institute, World Resources 1988-89 (New York, NY: Basic Books, Inc., 1988), p. 46.
128 • Energy in Developing Countries
Energy Production and Conversion
Energy plays a critical role in fueling the transi-
tion from a traditional to a modern society and thus
aiding economic and social development. In turn,
this energy is extracted from rural areas-either as
fossil fuels or renewable energy-and can have
significant impacts on the rural economy and
environment. Energy thus provides an important
two-way linkage between rural and urban areas.
Used wisely, energy can provide environmental
benefits; if misused it can also exact substantial
environmental costs to the land, water, and air. The
many social, economic, and environmental benefits
have been reviewed in the preceding chapters; the
environmental costs of extracting and transporting
the major energy resources used in developing
countries today-coal, oil, gas, hydroelectricity, and
biomass-are briefly reviewed below.
Coal
Coal has significant environmental effects through-
out the fuel cycle. 91 The environmental effects of
coal mining depend on the techniques used. Mining
methods are selected according to the depth of the
coal, the thickness of the seams, and the availability
of capital and equipment. In surface mining, topsoil
and overburden are removed to expose the coal. The
mining process can disturb surface lands and waters,
and may also contaminate or disturb underground
aquifers. Increased erosion, downstream siltation,
and water contamination can follow if excavated
material is not properly managed during mining.
Soil productivity and water resources can be de-
graded if lands are not reclaimed adequately. The
most severe impacts associated with underground
mining are surface subsidence over mined-out areas,
disruption of aquifers, and contamination of water
by acid drainage. Additionally, dust and emissions
from coal mining, preparation, and transport or
related equipment can contribute to local air pollu-
tion. 92
Oil and Gas
Oil and gas production have similar environ-
mental impacts. During exploration and production,
the major environmental concerns onshore include
land disturbances, aquifer contamination, leaks and
spills, and disposal of liquid and solid wastes. For
offshore operations, concerns include the impacts of
operations on fisheries and marine habitats, leaks
and spills, and waste disposal.
Large spills of oil or petroleum products, which
occasionally occur during production, storage, or
transportation, are perhaps the most dramatic envi-
ronmental threat associated with oil production.
Small leaks and spills are much more common,
however, and may have a greater overall impact on
the environment. Oil spills can poison fish and
aquatic animal and plant life. Additionally, sedi-
ments can trap oil, creating a long-term source of
pollution. The well-publicized Alaskan oil spill of
1989 revealed the difficulties inherent in cleaning up
large oil spills in difficult ocean environments. 93
Developing countries have had their share of oil
spills too. The J une 1979 blowout of the IXTOC 1
production well in the Mexican Caribbean was the
largest oil spill to date. 94 1 1 released an estimated 139
to 428 million gallons of crude oil into surrounding
waters, as much as 40 times the amount spilled by
the Exxon Valdez. Natural gas leaks pose little toxic
threat to plants and animals but can explode or cause
fires. The primary constituent of natural gas—
methane-is also a potent greenhouse gas.
Petroleum refining generates hazardous liquid
and solid wastes, as well as air emissions. These
facilities have pollution contrd requirements similar
to many other large industrial and chemical plants,
but leaks of toxic compounds occur nevertheless.
Hydroelectricity
The major environmental impacts of large hydro-
power production projects result from the initial
construction of the hydropower facilities, filling the
reservoirs, and changing river flows. This is in
"U.S. Congress, Office of Technology Assessment, The Direct Use of Coal, OTA-E-86 (Springfield, Y\: National Technical Information Service,
April 1979), p. 186.
^Ibid.
^For . di.c U ..io„ of the technologies related to ocean sp ju cleanup, see U.S. Congress, Office of Technology Assessment, Coping With An Oiled
Sea: An Analysis of Oil Spill Response Technologies, OTA-BP-O-63 (Washington, DC: U.S. Government Printing Office, March 1990).
^Ibid. Many of the large spills listed in the OTA background paper occurred in or near developing countries.
Chapter 5~Energy and the Environment in Developing Countries • 129
contrast with other energy sources, for which major
impacts are spread over the entire fuel cycle.
Large dam construction often requires the clear-
ing of lands for access routes and sometimes for
removal of construction material, with resulting soil
degradation and erosion. Filling the reservoir can
flood large tracts of land, uprooting people and
leading to loss of forests, wildlife habitat, and
species diversity. For example the Akosombo reser-
voir on theVolta in Ghana, with a land requirement
of 8,730 km 2 , approaches the size of such small
countries as Lebanon or Cyprus. 95 Some currently
proposed very large dam projects would inundate
highly populated valleys and require large-scale
resettlement of local residents.
The Three Gorges Dam project under considera-
tion for the Yangtze river in China would be the
world's largest hydroelectric project, generating the
equivalent of approximately one-fifth the hydroelec-
tricity currently produced in all of China each year,
as well as providing flood control. Debate over this
project has continued for six decades because it
would entail the resettlement of approximately 1
million people; because it would deface one of
China's most famous natural sites, and because
critics believe that power could be provided more
cheaply with smaller thermal and hydroelectric
plants or through conservation. 96
Dams disrupt the natural flow of rivers. Changed
flows can erode riverbeds, alter flood patterns, harm
aquatic ecosystems, and interrupt the spawning and
migratory patterns of fish and other species. The
introdudion of a new lake can affect water tables and
groundwater flows and interfere with the necessary
flow of nutrients, and may induce microclimatic
changes in humidity, cloud cover, and rainfall. In
addition, if the vegetation is not cleared from the
reservoir area before flooding, rotting organic matter
releases significant amounts of C0 2 and methane,
two potent greenhouse gases.
Hydropower facilities, in turn, are vulnerable to
environmental degradation. Heavier than expeded
si I tat ion of reservoirs from deforested and/or de-
graded lands upstream can reduce the lifespan of
hydroeledric projeds (see table 5-4). 97 The Hirakud
reservoir in India, for example, was expeded to be
productive for about 110 years, but now has an
estimated produdive lifetime of 35 years. 98 In Costa
Rica, excess sedimentation from soil eroded from
steep slopes planted with coffee trees over the past
20 years has caused estimated losses of $133 million
to $274 million at the Cachi hydroeledric station. 99
Dams and the irrigation schemes dependent on
these dams have also contributed to increased
incidence of debilitating diseases, such as schis-
tosomiasis. ""Smaller hydroelectric systems and
better management may minimize some of the above
adverse consequences.
Biomass
Biomass fuels, which are critical to the rural and
poor urban sedors of developing countries (chs. 2,
3, and 4), can have a variety of environmental
impads. These are discussed above and are closely
intertwined with deforestation, desertification, and
rural air quality. A later report from this study will
examine the environmental effeds of biomass use in
detail, particularly in the context of producing clean
biomass fuels (e.g., ethanol) and using them to fuel
modern equipment.
Nuclear Energy
Nuclear energy currently makes little contribution
to the overall energy requirements of developing
countries. Seven developing countries produce ura-
nium: South Africa, Niger, Gabon, India, Argentina,
Brazil, and Pakistan. 101 Eight developing countries
had operating commercial readors as of late 1989
and several more had commercial reactors under
95 R.S.Panday(cd.),Afan-marfe Lakes and Human Health (Paramaribo; University of Suriname, 1979).
^RobertDelfs, "wealth and Woe: The Long Struggle to Harness the Waters," Far Eastern Economic Review, vol. 147, Mar. 15, 1990, pp. 22-23;
She@ WuDunn, "In China, Dam's Delay Spares a Valley for Now," New York Times, Apr. 18, 1989, p. CI.
"U.S. Congress, Office of Technology Assessment, op. Cit., footnote 41, p. 43.
98 Ibid.,p. 43.
"David Dudenhoefer, "Forest Crisis Nears," The Tiw Times, vol. 34, Feb. 16, 1990.
1O0 United Nations Food and Agricultural Organization, op. cit., footnote 7, p. 69.
101 United Nations, Energy Statistics Yearbook, 1986 (New York, NY: 1988).
130 • Energy in Developing Countries
construction (ch. 4). A total of 28 developing
countries had research reactors as of late 1988. 102
The conventional nuclear fuel cycle includes
uranium mining and processing, fuel fabrication,
electricity generation, and radioactive waste dis-
posal. Each of these steps has the potential to release
varying amounts of toxic and/or radioactive materi-
als to the environment. Releases usually take the
form of small leakages but have, on rare occasion,
also resulted in catastrophes such as Chernobyl. On
the other hand, the nuclear power option releases
little carbon dioxide or other greenhouse gases to the
atmosphere, nor does it emit much sulfur dioxide,
nitrogen oxides, or other air pollutants. 103 These
tradeoffs pose difficult environmental choices.
Solar, Wind, and Other Renewable Energy
Solar energy can be used to heat water or dry
crops, or can be turned directly into electricity by
photovoltaic cells. Winds can be harnessed for
pumping water or generating electricity. Though
holding great promise for the provision of decentral-
ized forms of energy for remote areas, as yet these
sources provide only small amounts of energy for
developing countries.
Geothermal energy, though still a very small part
of total energy supply in developing countries, is
being used in several Latin American and Asian
developing countries and in Kenya. The environ-
mental problems associated with geothermal energy
production are highly dependent on the geochemical
characteristics of each specific site and the extent to
which good environmental practice is followed.
Potential adverse environmental impacts include the
release of gases (particularly CO 2 ), contamination of
local aquifers by saline (and sometimes toxic)
geothermal fluids into groundwater, subsidence of
land overlying wells from which geothermal fluids
have been extracted, and the generation of high-
temperature liquid effluents containing metals and
dissolved solids. With existing economically proven
technologies and good management, however, these
potentially adverse impacts can be kept under
control .
Greenhouse Gases and Global
Climate Change
The environmental impacts described above are
largely limited to the individual countries con-
cerned. Some activities- notably, the combustion of
fossil fuels and deforestation-can have a wider
impact, including impacts on the global climate. An
international panel of scientific experts of the I n-
tergovernmental Panel on Climate Change (IPCC) 104
recently concluded that: "emissions resulting from
human activities are substantially increasing the
atmospheric concentrations of the greenhouse gases:
carbon dioxide, methane, chlorofluorocarbons (CFCs)
and nitrous oxide. These increases will enhance the
natural greenhouse effect, resulting, on average, in
an additional warming of the Earth's surface. "105
Under the IPCC "Business as Usual" scenario,
emissions of carbon dioxide are projected to grow
from approximately 7 billion metric tonnes of car-
bon in 1985 to 11-15 billion metric tomes in
2025. 106 Methane emissions are projected to in-
crease from 300 million metric tonnes in 1985 to 500
million metric tonnes by 2025. Based on current
models and with these trends in greenhouse gas
emissions continuing, scientists predict that global
mean temperature will increase at a rate of about 0.3
C per decade during the next century, a rate higher
than that seen over the past 10,000 years. 107
102 World Resources Institute, World Resources 1990-1991 (New York, NY: Oxford University Press, 1990). Installed capacities are inadequate
indicators by themselves, as capacity factors for nuclear, thermal, and hydroelectric plants vary considerably. Installed capacities and capacity factors
can be found in WorldBank, "Summary Data Sheets of 1987 Power and Commercial Energy Statistics for 100 Developing Countries," Industry and
Energy Department Working Paper Energy Series No. 23, March 1990.
103 Some carbon dioxide may be released during minin g transport, r other operations by fossil-fueled equipment, and some maybe released in
fabricating the construction materials-particularly cement.
104 The Intergovernmental Panel on Climate Change (IPCC) is an international task force created by the World Meteorological Organization and the
United Nations Environment Program. Initiated in 1987, this body focuses on analyzing current information available on climate change issues and
devising strategies to address climate change.
^Intergovernmental Panel on Climate Change, "Policymakers' Summary f the ScientificAssessment of Climate Change: Report toIPCC From
Working Group I," June 1990, p. i.
'^Intergovernmental Panel on Climate a.»n g ., Working Group in (Response Strategies Working Group), "Policymakers' Summary of the
Formulation of Response Strategies: Report Prepared for WCC," June 1990, p. iii.
i°?IPCC Working Group I, op. cit., footnote 105, p. ii.
Chapter 5™Energy and the Environment in Developing Countries . 131
More than half of the warming has been attributed
to carbon dioxide. Atmospheric concentrations of
carbon dioxide have increased by about 25 percent
since preindustrial times. This rise is largely due to
carbon dioxide emissions from the burning of fossil
fuels and from deforestation.
I n 1985, according to estimates for the I PCC
Working Group III, developing countries contrib-
uted about one-quarter (26 percent) of annual global
energy sector Commissions; ™ three-fourths came
from the industrialized market countries and the
centrally planned European countries (including the
U.S.S.R.). By 2025, with expanding populations and
rapidly increasing energy use, developing countries
are projected to produce roughly 44 percent of global
energy sector C0 2 emissions.
While the Commissions from the energy sector
are fairly well known, there are large uncertainties
about the contribution of emissions from deforesta-
tion and other land use changes. This makes it
difficult to calculate with confidence the developing
country share of both total annual and cumulative
emissions for C0 2 and other gases. Estimates of the
Commissions from tropical deforestation differ by
a factor of four. 109 By various estimates, deforesta-
tion could be the source of between roughly 7 to 35
percent of total annual C0 2 emissions. Developing
countries are also responsible for at least half of the
global anthropogenic generation of two other impor-
tant greenhouse gases, methane and nitrogen oxides.
In addition to increases in mean global tempera-
ture, other effects expected to occur with global
warming would include increases in sea level 110 and
shifts in regional temperature, wind, rainfall, and
storm patterns. These, in turn, would submerge some
low-lying coastal areas and wetlands, threaten build-
ings and other structures in these areas, and increase
the salinity of coastal aquifers and estuaries. Such
changes could disrupt human communities and
aquatic and terrestrial ecosystems and affect food
production and water availability." A number of
developing countries may be especially vulnerable
to these effects. 112 No one, however, can yet predict
the timing, severity, or extent of the potential
impacts with certainty.
The I PCC scientists also noted that climate
models suggest that controlling emissions can slow
global warming, perhaps from 0.3 to 0.1 'C per
decade. Emission control strategies that countries
could consider today include improved energy
efficiency and cleaner energy sources-strategies
that may also have economic benefits. The expan-
sion of forested areas, improved livestock waste
management, altered use and formulation of fertiliz-
ers, and improved management of landfills and
wastewater treatment might also reduce or offset
emissions.
Achieving meaningful reductions in emissions
will require unprecedented levels of international
cooperation and must include developing countries.
In addition to the technological challenges for the
energy, agriculture, and industrial sectors, govern-
ments of the industrial and developing countries face
challenges in improving and expanding institutional
mechanisms for technology transfer to developing
countries for providing vital energy services while
limiting emissions. These issues are discussed in the
forthcoming OTA report Changing By Degrees:
Steps To Reduce Greenhouse Gases.
Government Policies
Responses to the deteriorating quality of the
environment in developing countries show a mixed
picture. On the one hand, progress is being made in
setting up the government apparatus of environ-
mental control, and concern and activity at the
10'IPCC Working Group m, op. cit, footnote 106, p. 10, table 2.
1<»IPCC, "Policymaker's summary of the Formulation of Response Strategies: ReporPrepared for IPCC by Working Group III," June 1990, p. 5.
Estimates of C0 2 emissions from deforestation and land use changes (including wood fuel) were 0.6 to 2.PtC in 1980 according to a report prepared
for the IPCC Working Group 1. "Scientific Assessment of Climate Change: Peer Reviewed Assessment for WG1 Plenary Meeting, May 1990," Apr.
30, 1990, p. 1-9. More recent evidence from selected countries would indicate deforestation rates now are higher than they werin 1980. Estimates of
CO, emissions in 1987 were 5.7 +0.5 PtC from fossil fuel burning and total emissions were of 7 +l.lPtC.
"The IPCC working group predicted an average rat. of global mean sea level rise of about 6 cm per decade over the next century ,20 cm by 2030,
and 65 cm by the end of the century with significant regional variations. This increase is primarily due to thermal expansion of theoceans and melting
of some land ice.
n 'Intergovernmental Panel on Climate Change, 'Policymakers' Summary f the Potential Impacts of Climate Change: Report from Working Group
nto the IPCC," May 1990, p. 8.
U2 J.D.Millimanet rd., "Environmental and Economic Implications of Rising Sea Level and Subsiding Deltas: The Nile and Bengal Examples,"
AMBIO, vol. 18, 1989, pp. 340-345.
132 • Energy in Developing Countries
grassroots la/el is rising rapidly. On the other hand,
implementation of environmental policies is fre-
quently difficult to achieve, and progress is some-
times impeded by unintended environmental im-
pacts of other policies.
The past two decades have seen considerable
environmental policymaking activity in the develop-
ing world, including the establishment of national
environmental agencies and organizations 113 and the
adoption of laws and practices designed to prevent
environmental pollution and degradation or to pro-
tect unique natural environments and wildlife. All in
all, more than 100 developing countries now have
governmental bodies whose responsibilities include
environmental protection. 114 However, only a limited
number of countries 115 have established independent
environmental agencies in their central govern-
ments. Most have increased their environmental
oversight capabilities by creating or expanding
sections or offices within environment- related gov-
ernment departments such as agriculture, energy,
fisheries, forestry, irrigation, natural resources, or
tourism.
The larger developing nations with traditions of
extensive government policymaking and implemen-
tation and a relatively advanced domestic research
base have formulated and adopted a variety of
environmental protection laws and standards, in-
cluding the requirement of environmental impact
statements for large projects. Developing countries
with detailed nationwide air and water quality
standards or with regulations specifying the levels of
permissible pesticide residues in food are still,
however, in the minority. Those that do have such
regulations have usually based their legislation on
the standards in force in the industrialized nations;
consequently, their specifications-at least on paper
—are usually as strict as and sometimes even stricter
than, those of the industrial countries, although
implementation may lag. 116
Some developing countries have shown a signifi-
cant commitment to protecting the environment.
China's most polluted coal-burning cities have mean
sulfur dioxide levels between 100 and 300 ug/m 3 ,
about the same as in British cities prior to the
introduction of Britain's Clean Air Act in the early
1950's. China, with levels of income much lower
than the United Kingdom in the early 1950's, has
already begun to respond to the problem by introduc-
ing cleaner urban fuels and outfitting coal-fired
plants with effective particulate controls 117 .
Environmental concerns in developing countries
have not been limited to domestic issues. Develop-
ing countries have also participated in international
environmental protection treaties. Most of the devel-
oping countries belong to the World Heritage,
Endangered Species, and Law of the Sea conven-
tions. Nine developing countries, including Mexico,
Egypt, Morocco, and Venezuela, have already
signed the Montreal protocol for protection of the
ozone layer.
Many developing countries have also experienced
an increase in grassroots environmentalism in addi-
tion to this official activity. In Ecuador, Colombia,
Guatemala, and Brazil movements organized by
citizen groups have led to the banning of toxic
pesticides and the initiation of major urban clean-
ups us indigenous nongovernmental organizations
(NGOs) have increasingly involved themselves in
environmentally oriented efforts. More than 600
Brazilian and Indonesian NGOs currently are work-
ing on environmental issues, as are several thousand
groups in India. 119
113 JeffieyH.LeonardandDavidMorell, , 'EmergenceofEnvironmentalConcerninDeveloping Countries: Apolitical Perspective," Stanford Journal
of International Law, Issue 2, Summer 1 981 .
114 Barbara J. Lausche, World Bank, "Environment and Natural Resources Management Institutions in Developing Countries,' background paper,
draft, Oct. 10, 1989.
115 Bermuda,BurkinaFaso,Gabon,India,Kenya, Oman, Papua New Guinea, Senegal, Singapore, Trinidad and Tobago, Venezuela, and Zaire.
1 16 The Chinese dairy average standard for ambient sulfur dioxide is stricter in each of the three classes (natural unpolluted areas, urban residential
locations, and industrial districts Hhan the U.S. values; maximum nitrogen oxide levels permissible in residential areas are equivalent to the Japanese
standard; and the maximum carbon monoxide levels allowed at any time are much lower than the 1-hour averages in West Germany or the United States.
T.A.Siddiqi and C.X. Zhang, "Ambient Air Quality Standard in China," Environmental Management, vol. 8, 1984, pp. 473-479.
117 Vaclav Smil, "E nvironment in Developing Countries, " contractor report prepared for theOffice of Technology Assessment, J une 1990.
118 WorldResources Institute, "Development and Pollution in the Third World," paper prepared for the Tokyo Conference on the Global Environment
and Human Response, Sept. 11-14, 1989, J uly 1989, p. 33.
119 Ibid.
Chapter 5™Energy and the Environment in Developing Countries • 133
Despite these achievements, major problems re-
main. A fundamental constraint on environmental
policymaking in all but a handful of developing
countries is the absence of systematic, integrated
data on the current state of environmental pollution
and degradation. While there is sometimes a great
deal of site-specific information connected with
particular projects, virtually all the developing
countries lack spatially representative networks for
monitoring land, water, and air quality or accurately
measuring the rates of deforestation or changes in
agricultural land and soil quality. China and India,
however, have accumulated a fairly extensive infor-
mation base through the combined activities of
various government departments and university-
based researchers.
The implementation and monitoring of environ-
mental policies and standards are also difficult. In
most countries, the budgets of the environmental
agencies are a small fraction of their minimum
needs, and staffing is inadequate. Enforcement is
often lax; prosecutions are often costly, uncertain,
and slow; and the punishments actually meted out
may be inadequate deterrents. 120
Environmental protection is also sometimes im-
peded-in both developing and industrialized coun-
tries-by government policies intended to promote
economic growth, generate employment, service
foreign debts, or meet other important national
needs. In Brazil, for example, a variety of govern-
ment policies have encouraged rapid development—
and thus rapid deforestation -of the Amazon. These
include tax exemptions, tax credits, subsidies of
rural credit, and land acquisition laws (squatter
rights). 121 These inducements have inadvertently en-
couraged a number of activities that would not be
economical in the absence of direct or indirect
government supports. For example, the implicit
government subsidy for ranching alone during the
period 1975 to 1986 has been estimated by the
World Bank at more than $1 billion. 122 The Brazilian
government has recently begun to reverse many of
these policies as awareness of their costs and
environmental impacts has grown.
Similarly, commercial logging has also been
encouraged by many governments through low
royalties and fees, reduced export taxes, and other
tax breaks for timber companies and other domestic
wood products industries. 123
Tax credits, investment subsidies, and other fiscal
supports may be necessary and justified to encour-
age needed development, and such policies are
widely used in developing and industrialized coun-
tries. The value of depletable natural resources and
the costs of environmental degradation, however,
are often not properly accounted for by these
financial instruments. For this reason, many econo-
mists advocate policies that more accurately include
these costs than has sometimes been done in the past.
With proper valuation of the natural resources,
environmental impacts, and other costs and benefits,
such investments might be made more wisely. 124
Government policies in some industrialized na-
tions may also inadvertently affect the environment
in developing countries. In 1988, the European
Community, the United States, and J apan provided
subsidies to their agricultural sectors totaling $97.5
billion, $67.2 billion, and $57.8 billion respectively—
a total of $222.5 billion. 125 This is nearly one-third of
I20 Critiques of government enforcement of environmental laws in developing countries can be found in: R. Abracosa, and L. Ortolano, 1987,
Environmental impact assessment in the Philippines: 1977-1985. Environmental Impact Assessment Review, vol. 7, pp. 293-310. CM. Abraham and
A. Rosencranz, 1986, "AnEvaluationof Pollution Control Legislation in India," Columbia Journal oj 'Environmental Law, vol. 11,1986, pp. 101-1 18;
H. Haeruman, "Conservation in Indonesia," AMBJO, vol. 17, 1988, pp. 218-222; A.D. Johns, Economic Development and Wildlife Conservation in
Brazilian Amazonia," AMBIO, vol. 17, 1988, pp. 302-306; J. Mayda, "Environmental Legislation in Developing Countries: Some Parameters and
Constraints," Ecology Law Quarterly, vol. 12, 1985, pp. 997-1 024; K. Ramakris hna, "The Emergence of Environmental Law in the Developing
Countries, a Case Study of India," Ecology Law Quarterly, vol. 12, 1985, pp. 907-935; A.S. Tolentino, "Legislative Response to Marine Threats in
the ASEAN Subregion," AMBIO, vol. 17, 1988, pp. 238-242.
121 The8° venlmentsus P CToed P rovisions fl^t made cl ear jrig for ranches profitable and penalized owners of unimproved lands, but ranches are still
attractive as a speculative hedge against inflation. See U.S. Congress, Office of Technology Assessment, op. cit., footnote 14; Climate Institute, "Climate
News Around the G\obe,' Climate Alert, vol. 2, No. 3, Fall 1 989, p. 8; S.B. Hecht, "The Sacred Cow in theGreenHell: Livestock and Forest Conversion
in the Brazilian Amazon," The Ecologist, vol. 19, No. 6, November/December 1989, pp. 229-234.
122 Hans P. Binswanger, World Bank, "Fiscal and Legal Incentives With Environmental Effects on the Brazilian Amazon," Agricultural and Rural
Development Department, Operational Policy Staff, May 1987; N. Myers, Deforestation Rates in Tropical Forests and Their Climatic Implications
(London: Friends of the Earth Limited, 1989).
123 Robert Repetto, "Deforestation in the Tropics,' Scientific American, vol. 262, No. 4, April 1990, pp. 36-42.
m llobert'R£pettDetai.,WastingAssets:NaturaIResourcesin theNationalIncomeAccounts(WisiaDgton,DC: World Resources Institute, June 1989).
125 Stuart Auerbach, "Statement Aimed at Farm Subsidies Impasse, « Washington Post, July 12, 1990, p. A9.
134 • Energy in Developing Countries
the GDP of the 2.8 billion people who live in
low-income developing countries, and it is compara-
ble to the GDP of their entire agricultural sector. 126
Some of these subsidies may tend to hold agricul-
tural prices down in developing countries as well as
limiting their export opportunities. 127 If this occurs
and reduces developing countries' earnings for their
agricultural output, they may have less incentive and
ability to invest in agricultural research and develop-
ment or infrastructure to achieve higher yields and
quality. As noted above, low agricultural productive-
ties (together with inequitable land tenure, etc.)
contribute to the expansion of croplands into ecol-
ogically fragile areas. Similarly, import tariffs to
protect domestic timber industries in industrial
countries may encourage inefficient harvesting in
tropical forests because the full market value of the
resources cannot be obtained. 128
However well-intentioned, the development poli-
cies of bilateral development agencies and multilat-
eral development banks have also sometimes di-
rectly contributed to environmental degradation due
to their emphasis on large projects-transporta-
tion infrastructures, hydroelectric facilities, and
industrial complexes. 129 Development agencies are,
however, becoming more sensitive to these environ-
mental concerns. 130 The role of development agen-
cies in promoting the adoption of more environmen-
tally sound energy technologies will be examined in
a later report of this OTA study.
126 WorldBank,VKjrWO«ve/o»mew/?e»<j«^^9(Washu^on,i)C:1989),in<Ucatortablesiand3.Notethattbis does not include the 1 billion people
in middle-income developing countries who had a 1987 GDP of nearly $2 trillion.
127 Vernon O. Roningen and Praveen m. Dixit, Economic Implications of Agricultural Policy Reforms in Industrial Market Economies (Washington,
DC: U.S. Department of Agriculture, Economic Research Service, 1989).
i28fj s Congress, Office of Technology Assessment, Op. cit., footnote 14.
,29 U.S. Congress, Office of Technology Assessment, op. cit., footnote 14; Bruce M. Rich, "The Multilateral Development Banks, Environmental
Policy, and the United States,"Eco/ogy Law Quarterly, vol. 12, No. 4, 1985; P. Aufderheideand B.Rich, "EnvironmentaIR eform and the Multinational
Banks," World Policy Journal, Spring 1988, pp. 301-321; W. Reid, "Sustainable Development: Lessons From Success," Environment, vol. 31, No.
4, May 1989, pp. 7-35,
130 See, f m -„i„ WorldBank, ' AReviewof the Treatment of Environmental Aspects of B@ Energy projects," Industry and Energy Department
Working Paper, Energy Series Paper No. 24, March 1990.
Appendix A
Glossary of Energy Units
Appendix A
Glossary of Energy Units
British thermal unit (Btu)--The basic unit of energy in
the English system of units. The energy required to
heat 1 pound of water 1 oF. One Btu is about the
energy given off by burning a single match tip.
Hectare (ha)-Anarea of land measuring 10,000 square
meters or equivalent^ 100 meters by 100 meters.
One hectare is equal to about 2.5 acres.
J oule (| )--A basic unit of energy in the metric system. A
joule of energy is very small (a single match tip
gives off about 1,000 joules), so joules are often
multiplied by orders often (10, 100, 1,000, etc.) to
yield useful units:
Name Value Abbreviation Equivalent
kilojoule thousand (10 3 ) joules kj 0.95 BtU
megajoule million (10*) joules MJ
gigajoule billion (10') joules GJ
terajoule 10" joules TJ
petajoule 10 1S joules PJ
exajoule 10" joules EJ 0.95 quads
Total energy production and consumption for a country is
typically given in exajoules; per capita energy consumption is
typically given in gigajoules (1 GJ is equivalent to the energy in
about 7.5 gallons of gasoline).
Kilogram (kg)--The basic unit of weight in the metric
system, equal to 2.2 pounds.
Kilowatt (kW)-The power (energy per unit time) unit in
which electricity is measured.
Kilowatthour (kWh)--The energy unit in which elec-
tricity is measured. A 100-watt light bulb burning
for 10 hours consumes 100 W X 10 hours= 1,000
Wh = 1 kWh of energy.
Meter (m)--The basic unit of length in the metric system,
equal to 39.4 inches.
Quad— 101 5 Btus. The United States currently consumes
about 80 quads (80X 10 15 Btus) of energy per year.
1 Quad is equal to about 1.05 exajoules (EJ ).
Tonne (t)--Short for metric tonne, equal to 1,000
kilograms or about 2,200 pOUfldS.
-137-